ESA title

Category: Philab

New ESA-GEOSAT deal to empower space solutions

Posted on by mavelar

ESA and GEOSAT have formalised their commitment to strengthening the space industry by signing a Letter of Intent, promoting entrepreneurship and advancing the development of innovative space solutions.

The latest company joining the portfolio of ESA’s Partnership Initiative for Commercialisation (EPIC) is the Portuguese company GEOSAT, one of Europe’s leading providers in Earth Observation (EO) satellite imagery and data analytics. GEOSAT provides very-high resolution (VHR) optical satellite data in Europe, being certified as a DPS Category 2 Provider (European Earth Observation Established Data Suppliers) and developing innovative EO products and services.

This partnership with GEOSAT will benefit companies supported by ESA Φ-lab, the InCubed Earth Observation commercialisation programme, ESA Phi-LabNET, ESA Business Incubation Centres (ESA BICs) and ESA Technology Brokers, operating under the broader ESA EPIC framework.

As an outcome of this Letter of Intent, GEOSAT will provide VHR data to ESA-supported start-ups, so they can test, validate and improve their services, along with technical mentorship and expertise, and joint outreach and networking activities to foster new opportunities and raise awareness about the societal value of Earth observation technologies. Further information on the collaboration and how to take advantage of the services provided by GEOSAT is now available here.

“By working with GEOSAT and opening access to their Earth observation constellation, we are allowing an easier access to data in order to improve solutions, in areas such as environmental monitoring, agriculture, infrastructure management, and maritime surveillance,” commented Michele Castorina, Head of the ESA Φ-lab Invest Office. “This collaboration not only benefits entrepreneurs and businesses across several sectors but also reinforces Europe’s competitiveness in the global space economy.”

To know more: GEOSAT, ESA Φ-lab, InCubed, ESA Phi-LabNET, ESA Business Incubation Centres, ESA Technology Brokers, EPIC

Photo courtesy of ESA

Earth observation indicators as the key to unlock the Global Goal on Adaptation framework

Posted on by mavelar

The 30th Conference of the Parties (COP30), held in Belém (Pará, Brazil), is strategically focused on implementation, with the challenge of creating a definitive, measurable framework for the Global Goal on Adaptation (GGA). Directly addressing this issue, a new ESA Φ-lab co-led article published in Nature offers a timely intervention, showing how satellite-based Earth observation data can provide the objective, globally consistent indicators needed to achieve climate resilience. 

10 November marked the start of the 30th Conference of the Parties (COP30) in Belém, Pará, Brazil’s gateway to the Amazon rainforest. Regarded as a key point in the global climate agenda, COP30 is shifting the focus from ambition to implementation and accountability, to make the Global Goal on Adaptation (GGA) framework finally operational.

Established by the 2015 Paris Agreement, GGA seeks to improve the ability to cope with climate impacts, build systems that can withstand shocks, and reduce susceptibility to climate hazards. With negotiations culminating on a final set of indicators to measure the progress towards this goal, the success of the summit will lie on implementing a scientifically robust and actionable framework.

Directly addressing GGA – and with a perfect timing – a new ESA Φ-lab co-led article, “Earth observations for climate adaptation: tracking progress towards the Global Goal on Adaptation through satellite-derived indicators”, has just been published in Nature.  This article is the result of the ‘Using Earth Observation Systems to Improve Climate Adaptation Policy and Action’ forum held last year in Bern (Switzerland), hosted by the International Space Science Institute (ISSI).

As part of the work performed within ESA’s Climate Change Initiative (CCI), “the paper highlights Earth Observation’s strengths in providing objective, repeatable, and globally consistent data, while also acknowledging challenges related to data disaggregation, integration with socio-economic factors, and the need for long-term, robust baselines”, the authors stated.

This research details how Earth observation (EO) data are relevant across the entire adaptation cycle, from initial risk to long-term monitoring and evaluation, focusing on four key sectors covered by the GGA framework.

For agriculture, satellites can monitor water-based variables such as evapotranspiration and soil moisture, as well as the status of surface water storages, the evolution of agricultural pests caused by climate change, and shifts in agro-climatic indices like aridity.

Regarding ecosystems and biodiversity, EO is uniquely positioned to measure the extent and changes of ecosystems like coastal mangroves, which serve as natural defences against sea-level rise and storm surges. Adaptation actions like those targeting the climate change-exacerbated threats of pests, droughts, and wildfires, are essential to protect the diverse environmental and socio-economic functions of forests, which include the provision of raw materials for bioeconomy, serving as a wildlife habitat, prevention of soil erosion, and facilitating carbon sequestration.

Extreme events such as floods, droughts, heat waves, or hurricanes, can be monitored by EO technologies, which provide insights across the different stages of the disaster risk management cycle, from pre-event assessment to post-event recovery. Additionally, EO data can be used to quantify vulnerability through detailed mapping of inhabited areas, building footprints, roads, and critical infrastructure such as dams.

EO data are instrumental in monitoring health-related hazards, in particular heat extremes, infectious diseases, and air pollution from wildfires, being frequently used as the input for models to produce hazard or exposure maps. While satellite-based data do not directly capture health outcomes, they are a proxy to perform health assessments at different levels: individual, household, cohort, or administrative.

Despite its immense potential, the authors stress that EO is not a solution on its own and provide recommendations for both the policy and scientific communities. They strongly urge negotiators to integrate EO data into the final GGA indicator toolbox that is set to be adopted at COP30.

Simultaneously, the authors call on the EO community to focus on how to operationalise EO-based adaptation data and information to make them easily accessible to policy makers. They also recommend a substantial investment in end-user training and embedding geospatial data science experts within operational agencies, especially in vulnerable regions, as this is essential for effectively using EO data for adaptation solutions.

The article concludes by highlighting that EO data must be integrated with socio-economic and local data to ensure it accurately reflects the context of adaptation and does not overlook the vulnerability of specific, often marginalised populations. 

Rochelle Schneider, Copernicus and Destination Earth Ecosystem Operations Engineer and the second author of the paper, commented: “To track real progress on adaptation, we need data that peacefully crosses borders effortlessly. This is exactly what EO satellites provide — globally harmonised evidence to support collective action against climate change impacts.”

This Nature perspective serves as the foundation for other efforts underway at ESA Φ-lab: Diego Jatobá dos Santos, an International Research Fellow supervised by Rochelle Schneider, is working on a project to assess the climate risks faced by children under different climate zones and climate change scenarios in Brazil, in a fruitful collaboration with UNICEF.

Diego will investigate an adjustment in UNICEF’s Children’s Climate Risk Index (CCRI) for Brazil climate zones, using diverse geospatial datasets and predicting CCRI under different climate change scenarios, with CMIP6 and/or Destination Earth Digital Twin data.

Building on this effort, Φ-lab is currently recruiting a Research Fellow to work on Artificial Intelligence (AI) for Climate Adaptation. The new lab member will investigate how AI can play a significant role in climate adaptation, resilience and mitigation.

The full Nature article is available here.

To know more: COP30, ESA Φ-lab, UNICEF’s Children’s Climate Risk Index (CCRI)

The banner image features forests around the Capim River (Rio Capim) in Brazil. Contains modified Copernicus Sentinel data (2022), processed by ESA.

Less is more: the power of TerraMind in your pocket and in space

Posted on by mavelar

On the last Earth Day, the European Space Agency (ESA) and IBM Research Europe launched TerraMind, a multimodal Earth observation foundation model. While powerful, the model’s need for large computing capabilities for fine-tuning and inference poses certain challenges, particularly in remote or resource-constrained environments. In October 2025, two lighter versions of TerraMind were released to be run on edge devices – like laptops or directly onboard satellites – enabling on-device, real-time analysis. The TerraMind Blue-Sky Challenge launched to even further improve TerraMind and has two new deadlines: 30 November 2025 and 31 January 2026.

Geospatial and Earth observation foundation models are crucial for overcoming global challenges. Pre-trained on vast, globally diverse datasets using self-supervised learning, these models learn a universal “language of Earth”, meaning that they can be quickly fine-tuned with minimal new data and supervised learning to perform a wide range of complex tasks, from disaster response and damage assessment to crop yield forecasting and wildfire spread forecasting.

This drastically reduces the cost, time, and data-labelling efforts required to create actionable insights, making complex Earth observation data more accessible for scientists, policymakers and communities worldwide.

As recalled during the last ESA-NASA International Workshop on AI Foundation Models for EO, the broader challenge now lies in moving beyond rapid-cycle prototyping toward the operational deployment of foundation models for real-world decision-making and societal benefits.

Achieving this requires designing models with deployment as a central consideration — ensuring they are efficient, accountable, and seamlessly interoperable with downstream systems such as digital twins, public dashboards, and early warning platforms. Enabling edge deployment, including onboard satellites, is vital for real-time analysis in resource-constrained environments.

To support this transition, building smaller but efficient models together with robust and scalable software stacks is essential to empower the broader community to adopt and apply these models effectively and responsibly.

In April 2025, ESA Φ-lab and IBM Research Europe launched TerraMind, a multimodal Earth observation foundation model. TerraMind is currently the best performing Earth observation foundation model, assessed by open community benchmarks such as PANGAEA, as it is the only model to outperform other traditional Earth observation methods with a frozen encoder strategy.

Six months later, two lighter versions of TerraMind – TerraMind.tiny and TerraMind.small – were released, specifically engineered to operate on normal or edge hardware devices. Despite their reduced size, these lightweight versions of TerraMind have little decrease in performance – compared with its full-scale version – allowing for model fine-tuning to any downstream use cases just with a standard computer, even without GPUs, or deploying the model to the ultimate edge, directly onboard orbiting satellites.

Performance of various TerraMind models (in blue) compared to other established foundation models (e.g., Prithvi 1.0, Prithvi-v2-300M), as assessed by the PANGAEA benchmark. mIoU (mean Intersection over Union) is a measure for performance in segmentation tasks. The higher mIoU value, the better a model performs.

Besides bringing Earth observation capabilities to edge devices, the creation of the ‘tiny’ and ‘small’ versions of TerraMind also lowers the entry barrier for Earth observation researchers, field workers, or smaller organisations. By drastically reducing the hardware requirements, anyone with a laptop can fine-tune these models and create new applications to better monitor Earth.

“The open-sourcing of the ‘tiny’ and ‘small’ TerraMind models is a game-changer for the entire Earth observation community. This is a clear step towards the true democratisation of planetary-scale foundation models,” commented Nicolas Longépé, Earth Observation Data Scientist at ESA Φ-lab.

“These lightweight versions shatter a barrier imposed by the lack of access to the latest, most powerful GPUs. They maintain industry-leading performance while running efficiently on edge devices and, most importantly, they move the power of cutting-edge AI from the exclusive domain of research centres and companies to the hands of the many who need it most,” Nicolas added.

Although the ‘tiny’ and ‘small’ versions of TerraMind represent a major leap in accessibility, innovation does not stop there: the TerraMind Blue-Sky Challenge, organised by ESA Φ-lab and IBM Research Europe, welcomes innovative ways to push TerraMind beyond “just another fine-tune”, whether it is prototyping a new multimodal workflow, exploring Thinking-in-Modalities, pushing the limits of these tiny but mighty models, or inventing a never-seen geospatial application.

This bi-monthly award has two new submission deadlines – 30 November 2025 and 31 January 2026, 23:59 (AoE). Each winner will receive a € 1.000 cash prize, with a potential publication in a joint wrap-up paper and broad visibility across IBM, ESA, and the Earth observation community. More information about this challenge can be found here.

All versions of TerraMind can be found here. TerraMind was developed within FAST-EO, an initiative led by a consortium comprising DLR, Forschungszentrum Jülich, IBM Research Europe and KP Labs, and supported and funded by ESA Φ-lab.

To know more: ESA and IBM collaborate on TerraMind, IBM ESA Geospatial @ Hugging Face, TerraMind Blue-Sky Challenge

Photo courtesy of Unsplash/Conny Schneider

Four new initiatives to boost Spain’s Earth observation sector

Posted on by mavelar

As the result of an ESA-dedicated commercialisation campaign for Spain, the InCubed Programme signed four new contracts with IVSEN, HAPSEYE, CrossBandInsights, and DVSTAI. From energy infrastructure monitoring to security and geospatial object detection, these projects reflect the growing impact of Earth observation data across key sectors.

Following the success of the last dedicated call for Spain, four initiatives signed a contract with the ESA InCubed programme. This call, launched in collaboration with the Spanish Space Agency (AEE), offered different levels of co-funding to develop innovative and commercially viable Earth observation products and services, while benefitting from the European Space Agency’s technical, commercial and financial guidance.

IVSEN is an advanced satellite-based monitoring solution tailored for energy infrastructure operators. It integrates a very-high-resolution payload (< 50 cm) with reduced mass and volume, along with agile observation modes for flexible operations. An on-board pre-processing algorithm works in tandem with the ground-based processing chain to generate specialised data products and analytics. This project is being developed by a consortium – SATLANTIS, Alén Space, DHV Technology, and GeoAI – with direct contributions from users such as Iberdrola to ensure the system meets real operational needs.

“IVSEN represents a strategic milestone for SATLANTIS, as it strengthens our capabilities in very high-resolution Earth observation — a core technology for the company’s future. We are grateful to ESA for their trust and support in driving this project forward, and for enabling us to deliver an agile solution that will help energy operators and other users monitor and safeguard their critical infrastructures,” stated Juan Tomas Hernani, CEO of SATLANTIS.

ICEYE delivers synthetic aperture radar (SAR) data worldwide through its fleet of satellites, supporting applications such as land use monitoring, border surveillance and environmental monitoring. To expand this capability, the company is developing HAPSEYE, a solar-powered, fixed-wing aircraft designed to operate at altitudes above 20 km for extended periods.

Equipped with a SAR payload, HAPSEYE will complement ICEYE’s satellite constellation by providing persistent, high-resolution imaging that overcomes current limitations in coverage and resolution. This next-generation platform will improve disaster response, security and environmental monitoring. The activity is planned to begin after the test campaign of HAPS Prototype-1, scheduled for late 2025.

“As a pioneer in SAR imaging radar satellite innovation, we are delighted to have been chosen for ESA’s InCubed programme in Spain. Initiatives like this are crucial for accelerating technological advancement and strengthening European competitiveness in the Earth Observation sector – objectives that resonate strongly with our mission at ICEYE. This commitment is underlined by the high-altitude platform station project we are taking on as part of the programme, designed to aid European natural disaster response and Earth Observation capabilities,” stated Lauri Väin, VP of High-Altitude Platforms at ICEYE.

TRE ALTAMIRA delivers satellite radar (SAR) displacement measurements and mapping solutions for sectors such as civil engineering, mining, oil, and gas. Its product, CrossBandInsights, enhances current single-frequency band interferometric SAR (InSAR) products, by combining X- and C-band observations with higher spatial and temporal observations to improve ground movement monitoring. This allows for engineering firms and authorities to detect subtle changes, supporting smarter infrastructure maintenance decisions and strengthening risk management with enhanced spatial and temporal coverage.

“InCubed Spain has given us the unique opportunity to turn our vision into a concrete product that will bring tangible benefits to the Earth observation market. CrossBandInsights addresses a critical need by merging multi-mission C-band and X-band InSAR data to provide more accurate and timely insights on ground deformation,” commented Roberto Montalti, Project Manager at TRE ALTAMIRA.  

“This innovation will support civil engineering companies and public authorities in ensuring infrastructure safety and resilience. We see this project as a clear example of how public funding can be effectively invested to foster innovation, create market-ready solutions, and strengthen Europe’s position in the space sector,” Roberto added.

Thales Alenia Space, a global leader in space manufacturing, has been delivering advanced solutions in telecommunications, navigation, Earth observation, environmental management, science and orbital infrastructures for over 40 years. Among its innovations is DVSTAI (Deeper Vision Self-Trained AI), an evolution of the SatHound project, designed to overcome the limitations of current geospatial object detection methods: traditional approaches often require expert intervention for model design, training, and deployment, making the process slow, costly, and vulnerable to risks such as unauthorised access or data leakage.

DVSTAI addresses these challenges by leveraging deep learning techniques, allowing even non-AI or non-Earth observation specialists to autonomously train and use models through a user-friendly software solution. These models can be tailored to specific applications, including object detection, change detection, and semantic segmentation.

“DVSTAI is a user-centric AI solution that empowers non-technical users to autonomously create, train, and deploy AI models for object detection and vision tasks over satellite imagery. It simplifies the process, reduces costs, and enhances security by eliminating the need for dedicated AI engineers to develop high performing vision models, making it an invaluable tool for EO analysts and service providers,” commented Julian Cobos, Product Line Manager at Thales Alenia Space Spain.

“Thanks to the ESA InCubed programme, Thales Alenia Space will develop new key capabilities for object detection in Very High Resolution (VHR) and Synthetic Aperture Radar (SAR) data and bring DVSTAI solution to the public Cloud in a Software as a Service (SaaS) model, making it accessible for any user to set-up object detection campaigns over open and commercial data sources,” Julian added.

To know more: ESA InCubed, Spanish Space Agency (AEE)

Photo courtesy of Unsplash/Chris Boland

ESA InCubed and UKSA fund five Earth Observation projects

Posted on by mavelar

The joint ESA InCubed/UKSA funding call has awarded over £ 2.5 million to five projects that will turn Earth Observation data into essential public services. The funding supports projects tackling national priorities: CORE for safer infrastructure monitoring, GHGSat’s platform for tracking methane emissions, and three systems – THICKET, FANTOM and EO4Biodiversity – designed to support sustainable land management and enhance biodiversity.

The European Space Agency (ESA) and the UK Space Agency (UKSA) have announced the results of their joint InCubed funding call, awarding over £ 2.5 million to five innovative projects that use satellite data to improve public services.

This initiative is a clear example of how the ESA InCubed programme supports its member states’ governments in the development of a domestic space industry that serves public good. The call’s explicit requirement for proposals to target a public sector end-user shows that ESA is actively steering its investment towards applications that can directly benefit citizens and government operations.

After a very successful and fierce competition, here are the new five ESA InCubed/UKSA-funded projects:

CORE: satellite insights for infrastructure safety

Corner Reflector Enabled Remote sensing (CORE), developed by Geospatial Ventures Limited and Bloc Digital, is a solution for monitoring public infrastructure and ground stability.

Traditional surveying methods are often costly, slow, and pose a risk to personnel, especially when inspecting large building complexes or difficult-to-access terrain. CORE addresses these challenges by combining multiple satellite data streams – from Interferometric Synthetic Aperture Radar (InSAR) and high-resolution optical imagery – to detect small movements, providing a much clearer, more comprehensive view of conditions across urban and rural landscapes than single-source systems.

CORE will translate complex satellite data into actionable intelligence for public sector users, such as engineers, urban planners, and environmental managers. By integrating satellite data with artificial intelligence and machine learning, the system provides early warning of ground shifts, structural settlement, or degradation before issues escalate into dangerous or expensive failures.

“Through CORE, we’re demonstrating how Earth observation—InSAR and optical—can deliver valuable and practical benefits for stakeholders by providing early insights into environmental change, ground stability, and asset condition. ESA’s support through InCubed is essential in helping us accelerate development, integrate advanced EO capabilities, and transform satellite intelligence into actionable information that helps organisations manage risk, reduce maintenance costs, and strengthen environmental resilience,” stated Paul Bhatia, Managing Director at Geospatial Ventures Ltd.

THICKET: a biodiversity mapping tool to support sustainable agriculture

THICKET is a tool being developed by AAC Clyde Space to help farmers enhance sustainability and better support wildlife on their lands.

The system will use the upcoming VIREON constellation of satellites, which will capture frequent, high-resolution multispectral images – with a detailed 1.5-meter resolution – to map habitats across farmlands. This constellation was engineered to provide well-aligned Earth observation data, including spectral bands that align with Sentinel-2 bands, complementing existing initiatives like Copernicus.

By showing farmers what biodiversity assets they have, THICKET provides the data for them to make informed, sustainable farming decisions. This capability is crucial for supporting environmental management and directly helps farmers meet the requirements to access valuable government support programmes like the Sustainable Farming Incentive.

“The ESA InCubed/UKSA co-funding has been instrumental in accelerating the development of THICKET. By combining advanced satellite technology with AI, we are creating a scalable, cost-effective way to monitor biodiversity across farmland. With imagery from our upcoming VIREON constellation, we can now capture fine details like hedgerows and flower margins — features that were previously almost impossible to assess systematically. This collaboration is helping to make biodiversity visible, measurable and actionable, supporting both farmers and the UK’s broader sustainability goals,” commented Pamela Smith, Director of Government Programmes at AAC Clyde Space.

Government GHG Service: tracking methane for net-zero

GHGSat UK and Terrabotics are developing an advanced analytics platform, Emissions Watch Service, to convert satellite observations of greenhouse gases (GHGs) into practical, actionable insights for the government. This service is uniquely positioned to support the UK’s goals of reaching net-zero emissions by 2050. By using their proprietary constellation of satellites, GHGSat traces the source of GHG emissions directly to specific industrial facilities, with a focus on the powerful GHG methane.

The platform enriches raw satellite data with detailed industrial asset information, creating a robust tool for environment compliance and reporting. Its rapid detection capability will ensure accurate data is available within hours of an emission event, allowing the UK government to make informed decisions about targeted mitigation strategies and increase accountability across major emitting organisations.

“ESA InCubed is a powerful programme, harnessing the innovation of space-based technologies for government agencies in the UK and Europe. For GHGSat, the support from InCubed is critical in order to de-risk product development while leveraging ESA’s technical expertise, enabling us to develop a platform that is honed to solve the unique challenges that government faces,” commented Daniel Wicks, Managing Director at GHGSat UK.

“Ultimately, through InCubed, GHGSat will strengthen its analytics prowess, identifying and mapping sources of methane to create a comprehensive view of emissions to inform data-backed policy, strengthen regulatory compliance, and drive methane reduction,” Daniel added.

FANTOM: advanced environmental analytics for land management

FANTOM (Future Analytics for Nature Through Observation and Modelling) is a project from Earth-i and Specto Natura designed to transform environmental land monitoring across the UK. It builds a database of agricultural and biodiversity markers, creating novel environmental indicators directly from satellite imagery.

FANTOM’s scope will extend well beyond agricultural subsidies: the platform is designed to provide content and context that supports not only agricultural schemes but also net zero and broader climate change mitigation activities. The comprehensive database of markers and impact assessments will be made available to all governmental agencies, associated arms-length bodies and commercial companies, enabling them to monitor and measure the progress of their sustainability activities and interventions.

“Earth-i’s FANTOM project, supported by the InCubed programme, will build a high spatial and temporal resolution, UK-centric database of agricultural and biodiversity markers with rich information content,” comments Jennifer King, Project Manager at Earth-i. 

“This will support environmental schemes aligned with the UK’s Agricultural Transition and assist government policy implementations for Net Zero and biodiversity net gain. FANTOM will provide analytics directly to the Rural Payments Agency, which manages farming subsidies and environmental schemes for England. Following this, Earth-i aims to promote the service to other countries, tailoring the analytics products as necessary,” Jennifer added.

EO4Biodiversity: satellite tracking for habitat net gain

EO4Biodiversity is an innovative project led by HR Wallingford to leverage satellite data to improve plant and animal diversity. The project’s aim is to automate biodiversity impact assessments by developing new ways of using Earth observation data to track habitat changes over time.

EO4Biodiversity will streamline the assessment process for land development and environmental management by post-processing existing Earth observation datasets, such as the ones from ESA WorldCover, specifically for biodiversity evaluations. By automating complex assessments, this initiative moves beyond manual surveying, providing public entities and other stakeholders with a powerful and scalable service to inform planning decisions, ensure compliance, and strategically support the long-term sustainability of the UK’s natural environment.

“EO4Biodiversity is a UK-wide project that uses satellite data to improve how we measure the impact of infrastructure projects on nature. With support from InCubed, the team is developing a new tool that will help landowners, developers, and public organisations understand how different building plans affect local biodiversity. This will make it easier to protect and enhance natural habitats while meeting the UK government’s biodiversity net gain targets,” stated Marta Roca Collell, Principal Engineer, Flood and Water Management, HR Wallingford.

The campaign manager, Pejman Nejadi (End-to-end Systems Engineer at the ESA Φ-lab Invest Office), commented: “This campaign stands as a clear demonstration of the value that ESA’s InCubed programme can deliver in partnership with national agencies. By combining ESA’s unique technical expertise and programme management experience, with UKSA’s strong understanding of national priorities, we created an initiative that directly addressed the UK public sector’s need of Earth Observation data. The success of this campaign highlights both the strength of our collaboration and the effectiveness of InCubed in fostering solutions that bring real benefit to society.”

To know more: ESA InCubed, Geospatial Ventures Limited, Bloc Digital, AAC Clyde Space, GHGSat UK, Terrabotics, Earth-i, Specto Natura, HR Wallingford

Photo courtesy of Unsplash/Paul Fiedler

AI challenge advances satellite-based disaster mapping

Posted on by mavelar

Four teams from different countries have been recognised for their breakthrough work in using artificial intelligence to detect earthquake damage from space, marking the conclusion of a global competition organised by the European Space Agency in collaboration with the International Charter ‘Space and Major Disasters’ – commonly referred to as ‘the Charter’.

The winning teams – TelePIX from the Republic of Korea, Datalayer from Belgium, DisasterM3 from Japan and Thales Services Numériques from France – were honoured recently during a ceremony held at the Charter’s 54th Board Meeting in Strasbourg, as France’s French Space Agency, CNES, took leadership of the Charter for the next six months.

Combining the Charter’s operational experience with ESA Φ-lab’s drive for innovation, the ‘AI for Earthquake Response Challenge’, which is part of the ESA Φ-lab Challenges initiative, brought together 143 participants from 40 countries to explore how far artificial intelligence can go in automating post-disaster damage detection from space.

Read the full article on www.esa.int.

Photo courtesy of ESA Φ-lab Challenges.

The BiDS Award spotlights top European space start-ups

Posted on by mavelar

The BiDS Award, a joint initiative by ESA Φ-lab and ESA BIC Latvia, took place at the 2025 Big Data from Space (BiDS) conference to boost European space start-ups and academia. The award provided visibility and networking opportunities for those working on space-based solutions that address global challenges, facilitating the commercialisation of deep-tech innovations. The winners of the 2025 edition were AgroRisk, SALUTS, and Hyphorest.  

The BiDS Award, organised jointly by ESA Φ-lab and ESA BIC Latvia, took place on 2 October 2025, within the framework of the Big Data from Space (BiDS) conference. This award brought industry, research, and policy leaders together to explore how deep tech solutions and space data can be leveraged to address critical global challenges.

This initiative focused on transforming raw data into knowledge, insight, and foresight, showcasing how advancements on space technologies are boosting data usage to deliver impactful planetary solutions, and providing visibility and networking opportunities for European space start-ups and academia.

While proposals at the intersection of Earth observation and global challenges were central, the scope was broad, encouraging solutions from across the entire space value chain. This included upstream, downstream, and spin-in innovations such as advanced materials, robotics, quantum technologies, satellite communications, in-orbit services, navigation, and AI-driven analytics.

The award’s areas of focus included precision agriculture and food security, environmental sustainability and biodiversity tracking, or sustainable energy and infrastructure monitoring, among many more. The winners were awarded one year of free access to AI-data analytics platforms from Altair (in a total value of € 300.000) and access to satellite imagery from Airbus (valued in a total of € 25.000).

The third place was awarded to Hyphorest (Germany), a Stuttgart-based startup and incubatee of ESA BIC Baden-Württemberg that aims to bring finance to nature by making it easy to invest in restoration, carbon farming, and carbon removal projects that are measurable, transparent, and grounded. Using satellite data, AI, and blockchain, Hyphorest quantifies natural impact such as biomass, biodiversity, CO₂ storage, and ecosystem recovery, while strengthening the role of local and indigenous communities in achieving these goals.

The platform enables companies and individuals to invest in climate and nature-positive projects with measurable outcomes, turning environmental and social impact into a trusted, data-driven asset class. The solution presented at BiDS helps companies invest in nature, track their impact, and report results within their corporate sustainability targets and frameworks. Hyphorest’s mission is to build trust in nature-based investments and make real progress visible to everyone.

“Winning the BiDS Award affirms our belief that space technology is one of the most powerful tools for climate resilience. At Hyphorest, we turn satellite data into living evidence of restoration, making environmental impact measurable, transparent, and investable. Being recognised by ESA Φ-lab and the BiDS community encourages us to keep pushing the frontier where space innovation meets nature,” commented Hojjat Mansourpour, Founder and Chief Executive Officer of Hyphorest.

SALUTS (Germany) confirmed its position as a leading innovator in the European space-tech ecosystem by winning the second place. SALUTS’ mission is to redefine AI-driven autonomy in space and beyond by transforming space computing with ultra-efficient chips that enhance real-time, in-orbit data processing. Their vision was first recognized in 2023 when SALUTS won the 5th CASSINI Hackathon, an achievement that led to their selection by ESA BIC Bavaria to further develop their winning project.

The innovation that secured their BiDS Award recognition is CHIRB (Computing on Hybrid Interplanetary Relay Basis), a revolutionary advanced AI system that acts as the next-generation mission control centre for space, defence, and industrial AI applications. At the heart of CHIRB is Robot-on-a-Chip, SALUTS’ proprietary AI technology, whose chips are engineered to use 90% less power while providing three times more reliable data processing directly on the device.

CHIRB integrates seamlessly with existing infrastructure, featuring a natural language interface that lets any user simply chat their requests, which are converted by the system into working code and hardware control, thus eliminating the need for complex coding or technical knowledge.

“At SALUTS, ‘We Deliver Autonomy at the Edge and Clarity at Scale’. Our platform, CHIRB, combines ultra-low-power hardware modules with advanced AI software to make complex operations autonomous, efficient, and sustainable. The outcome: faster decisions, lower costs, and more reliable operations,” stated Mohamed Sobhy Fouda, Chief Executive Officer and Founder of SALUTS.

The first place was awarded to AgroRisk (Denmark), a climate fintech platform and incubatee of ESA BIC Denmark that quantifies agricultural and financial risks caused by climate change and extreme weather. By combining Earth observation data, weather models, and financial risk analytics, AgroRisk enables banks, insurers, and farmers to assess the climate exposure of agricultural assets at both field and portfolio level.

The platform translates satellite and climate data into actionable financial insights, helping financial institutions comply with new sustainability regulations and supporting farmers in adapting to a changing climate. AgroRisk contributes to food security, climate adaptation, and sustainable finance — leveraging space technology to enable data-driven resilience in the agricultural sector.

“Space data is transforming how we understand and manage climate risks on Earth. At AgroRisk, we use satellite-based insights to translate what happens in orbit into tangible value on the ground — helping banks, insurers, and farmers make smarter, more sustainable decisions. The BiDS Award highlights how space innovation can directly contribute to planetary resilience and sustainable finance,” commented Theodor Christensen, CEO and Founder of AgroRisk.

The winners of the 2025 BiDS Award. From left to right: Sabrina Ricci (AI Ecosystem Coordinator at ESA Φ-lab), Hojjat Mansourpour (CEO and Founder of Hyphorest, 3rd place), Mohamed Sobhy Fouda (CEO and Founder of SALUTS, 2nd place), Theodor Christensen (CEO and Founder of AgroRisk, 1st place), and José Manuel Delgado Blasco (Geospatial System Engineer at ESA Φ-lab). Photo courtesy of ESA/Sabrina Ricci.

A special recognition goes to Andrii Chepurnyi, PhD student at the University of Latvia, who received a prize for his Earth observation-calibration project. As part of the award, he will join the Commercialization Reactor’s Commercialization Dive programme.

Andrii Chepurnyi (middle), PhD student at the University of Latvia, won a special recognition. Photo courtesy of Gatis Orlickis.

“The BiDS Award is another example of collaboration between organisations dedicated to developing the European Space capabilities in Deep Space and Earth Observation. In this occasion, both industry and academia responded positively, as well as partners and sponsors needed to make this award a success. This success, along other parallel efforts, wants to push the limits of Space technology and help develop European champions that proposed solutions to real problems with space technology,” commented José Manuel Delgado Blasco, Geospatial System Engineer at ESA Φ-lab and co-organiser of the BiDS Award.

“In this award, we have collected many brilliant participants addressing problems such as space debris, environmental and atmospheric pollution, and climate change. It has been a pleasure to work with ESA BIC Latvia, Airbus and Altair and I want to thank all the people involved and participants who made this award a success – and a starting point for future collaborations,” José added.

“The success of the BiDS Awards within such a vibrant and diverse ecosystem as the Big Data from Space community confirms how rewarding it is to push the boundaries of space and connect with emerging pioneers of technology. Their contributions help us drive innovation both in Earth Observation and in Space. The enthusiastic response from sponsors, participating companies — including those who did not win — has encouraged us to continue promoting these initiatives as catalysts for new collaborations and interactions,” commented Sabrina Ricci, AI Ecosystem Coordinator at ESA Φ-lab and co-organiser of the BiDS Award.

To know more: BiDS Award, ESA Φ-lab, ESA BIC Latvia

Photo courtesy of Gatis Orlickis

Towards a ‘Mission Control for Earth’: Better understanding Earth’s systems using AI and space data

Posted on by mavelar

In August 2025, FDL Earth Systems Lab presented three big AI research outcomes to improve how we understand and predict Earth’s changing systems and offer a window on how we might build a ‘Mission Control for Earth’. Leveraging the European Space Agency’s missions and funded by ESA Φ-lab, this initiative combines fresh datasets with innovative AI tools to give the global community better ways to track and respond to our planet’s most urgent environmental shifts.

“Guided by artificial intelligence, driven by human good”. This could be FDL Earth Systems Lab (ESL)’s motto. ESL is a research collaboration framework funded by ESA Φ-lab and implemented by Trillium Technologies, with the support of University of Oxford, Google Cloud, NVIDIA, Scan AI, and Pasteur ISI. It focuses on artificial intelligence (AI) – in particular machine learning (ML) – to support Earth sciences, helping researchers create practical tools for some of humanity’s toughest challenges with the best of motivations: ‘planetary stewardship’. 

FDL Earth Systems Lab has run annually since 2008. Experts with deep knowledge of the challenge domain work side by side with data scientists to develop new AI-enhanced approaches and tools. The short, focused format encourages quick testing and refinement, ensuring stronger results.

Last August, the ESL 2025 Live Showcase featured three ambitious research sprints: (1) refining 3D cloud models to improve forecasts of extreme events; (2) testing how well foundation models perform in sparsely observed events; and (3) advancing onboard ML to spot short-lived atmospheric events, such as greenhouse gas emissions. Each sprint brought together unique datasets and new AI-based methods to support the global research community.

Photo courtesy of Trillium Technologies.

3D CLOUDS FOR CLIMATE EXTREMES

Advancing global 3D cloud reconstruction is essential to deepen our understanding of cloud structure and the interactions with terrestrial and atmospheric phenomena. This is critical for tropical cyclones, which remain among the hardest weather systems to predict, especially during the intensification stage. Forecasts often poorly resolve a cyclone’s internal dynamics, simulations of cloud properties are highly uncertain, and observational records are limited, with only about 80 to 90 tropical cyclones occurring each year. The ‘3D Clouds for Climate Extremes’ sprint builds on a mature model training pipeline established in ESL 2024, which successfully modelled 3D clouds from geostationary data.

First, the team pre-trained a sensor-independent model on a large dataset of top-view satellite imagery from GOES-16, MSG and Himawari-8, to reconstruct masked versions of the observations. Second, they fine-tuned the model using a dataset from CloudSat, which provides vertical cloud profiles. The team also created a benchmark dataset, by combining satellite imagery with the timing and location of cyclone events. Since the model is sensor-independent, it is possible to include other satellite data that were not used for training, ensuring global coverage.

Together, these data enable the reconstruction of key microphysical properties of clouds, including ice water content (notably elevated in rapidly intensifying cyclones), droplet effective radius (a critical factor in cloud absorption and reflection of sunlight), and radar reflectivity (linked to cloud density and an indicator of rainfall).

Improving the prediction of cloud structures in three dimensions opens opportunities for a wide range of scientific and applied use cases: forecasts of hurricane intensity, discriminative cloud classification, or to understand how deforestation influences cloud cover and type. This ambition aligns closely with the objectives of ESA’s cloud, aerosol and radiation explorer mission,EarthCARE, which aims to advance our understanding of cloud-aerosol-radiation interactions.

FOUNDATION MODELS IN EXTREME ENVIRONMENTS

Earth observation foundation models are very powerful tools, but they also have limitations, especially when facing unfamiliar scenarios such as extreme events. One of the reasons is that training datasets typically contain limited examples from these events, leading to weaker performances when applied outside the conditions represented in the data.

When queried about a particular topic, foundation models can be ‘confidently wrong’. This becomes especially problematic when these models are used in critical, time-sensitive situations such as disaster response. It is essential to increase model transparency in cases where the model output has a high degree of uncertainty and requires human validation. But how can we know if the model is uncertain? 

The ‘Foundation Models for Extreme Environments’ team brought a novel answer to that question. The team – mentored by Φ-lab’s Internal Research Fellows Patrick Ebel and Ruben Cartuyvels – focused on distinguishing two types of uncertainty: data-driven or model-driven.

SHRUG-FM (Systematic Handling of Real-world Uncertainty for Geospatial Foundation Models) was developed as an adaptable framework for the community that combines input and training image comparison, embedding comparison, and the foundation model’s output and uncertainty into a planning and selective prediction mechanism, to ensure that the model can give a prediction, raise a warning, or simply say that it does not know the answer.

STARCOP2.0: ATMOSPHERIC ANOMALY DETECTION FROM ONBOARD

One of the most urgent applications of Earth observation is detecting and tracking greenhouse gas (GHG) emissions that are driving global warming. Methane, in particular, is one of the most powerful heat-trapping gases. Hyperspectral satellites play a crucial role in the detection of such gases: each gas interacts with light in a unique way, creating a distinct ‘spectral signature’ or ‘fingerprint’ that allows its identification from space.

The STARCOP 2.0 solution is built on a ‘tip-and-cue’ system that makes use of hyperspectral satellite data. In this setup, the ‘tip’ satellite is responsible for quickly detecting methane plumes. Once a plume is identified, it alerts the ‘cue’ satellite, which carries out more advanced tasks such as detailed plume segmentation and estimating methane concentrations using a U-Net ML model.

Unlike traditional approaches, image analysis happens directly onboard, avoiding delays from sending images to ground stations for processing. To achieve this, the team built two ML-ready datasets, one with orthorectified images, and another with un-orthorectified images that are more suitable and realistic for onboard implementation. These datasets were used to train three models, bypassing the need for image correction and reducing inference time.

The datasets have been shared with the community, and the models are being optimised for spacecraft limitations in computing power, memory and energy. This system makes it possible to detect methane and other GHG leaks quickly, helping policymakers hold polluters accountable and support efforts to reduce emissions.

“We’re motivated to show how AI’s powerful predictive and insight-extracting toolbox can make a significant difference to how we monitor and manage our planet. What’s exciting about this year’s research products is that we are showing how multi-instrument methods and context-aware AI can be harnessed to make a dent in open problems – such as rapidly determining the anatomy of a cyclone or identifying erroneous greenhouse gas emissions from orbit. If you are a tech optimist – which we are – you will see that the puzzle pieces for a ‘mission control for Earth’ are now within our reach,” commented James Parr, Founder and Chief Executive Officer at Trillium Technologies.  

Nicolas Longépé, Earth Observation Data Scientist at Φ-lab, is ESA’s Technical Officer for the initiative: “The FDL sprint format works because it brings together experts from different fields to collaborate intensively and prototype solutions quickly. By combining domain specialists, AI researchers, and technical mentors, we can tackle complex, carefully chosen challenges with real impact. These three sprints fit perfectly into the Earth Action paradigm we pursue at Φ-lab, moving beyond passive observation towards proactive insights and decision-making for a more resilient planet.”

To know more: ESA Φ-lab, Trillium Technologies, FDL ESL AI SOTA Live Showcase

Photo courtesy of Trillium Technologies

Advancing AI for Earth observation at the REO workshop

Posted on by mavelar

The first ‘REO: Advances in Representation Learning for Earth Observation’ workshop will bring together researchers and practitioners from machine learning, computer vision, and Earth sciences to advance the development of robust, interpretable, and scalable models for monitoring our planet. The ‘Call for Papers’ submission deadline is 20 October 2025.

(Updated on 15 October 2025)

Taking place at the Bella Center Copenhagen on 6/7 December 2025, the Representation Learning for Earth Observation (REO) workshop – part of EurIPS, a European conference officially endorsed by NeurIPS – will gather experts from machine learning, computer vision, and Earth sciences to present the latest research, discuss real-world scientific uses, and share innovative system designs.

With massive, diverse datasets from satellites and other sensors becoming widely available – and with the rise of general-purpose foundation models – Earth observation faces new opportunities and complex challenges. But how can we best combine these various streams of information to create useful applications?

The development of representation learning algorithms that understand raw Earth observation data with minimal human instruction is gaining traction beyond university labs. This interest is highlighted by projects from technology leaders such as Google DeepMind’s AlphaEarth, ESA-IBM’s TerraMind, AllenAI’s Earth System, or Meta’s DINOv3.

This growth calls for more focused discussions on how to develop, deploy, and use these powerful models. The workshop will address fundamental questions like “Where is the field today, and what steps should the community take next?”, “What are the biggest hurdles in getting computers to effectively learn from Earth data?” or, given the trend towards general-purpose, one-for-all AI models, “What is the role of specialised approaches for Earth science?”

Participants are invited to present their novel work as extended abstracts or discuss recently published work that is relevant to the workshop. The ‘Call for Papers’ submission deadline is 20 October 2025. While the current deadline is set, organisers advise potential contributors to check the workshop’s website for any possible updates.

The topics are broad and exciting, including new approaches in machine learning for Earth observation, such as self-supervised, multimodal, and domain-adaptive models. Experts will discuss the combination of AI with physics, and the integration of established models into AI pipelines to get better predictions and understand the uncertainty in their results.

A major focus is on ecology and environmental monitoring, covering essential tasks like tracking changes in land use, mapping biodiversity, estimating forest biomass, and assessing the conditions of soil and vegetation.

Technical discussions will also focus on the difficulties of remote sensing data processing, such as combining different types of sensors and ensuring consistency between different satellites.

Discussions on data curation and accessibility will cover how to build fair, accurate, and easily accessible global datasets for research, ultimately driving real-world innovations in applications like mapping urban areas or monitoring natural disasters.

Leading scientists and industry experts will give keynote presentations: Gustau Camps-Valls from the IPL lab of the University of Valencia, Michal Kazmierski from Google DeepMind, Julia Gottfriedsen from OroraTech, Bertrand Le Saux from the European Commission, and Ankit Kariryaa from the University of Copenhagen.

“REO will provide an amazing opportunity for machine learning researchers and practitioners that are interested in Earth observation to find each other in Europe,” commented Ruben Cartuyvels, Internal Research Fellow at ESA Φ-lab. “The current trend in AI4EO of representation learning with foundation models of increasing size leaves many open questions, and fruitful community exchange is essential to take steps towards answering those”.

Interested parties can find out more about this workshop and submit their abstract here.

This workshop is being co-organised by researchers from ESA Φ-lab, the University of Copenhagen, the Technical University of Berlin, IBM Research Europe and ENPC.

To know more: REO Workshop

The banner image contains modified Copernicus Sentinel data (2024), processed by ESA

Leadership and technology: AEE boosts Spain’s resilience against wildfires

Posted on by mavelar

Given the scale of the recent wildfires that have significantly affected Spain’s land and ecosystems, the Spanish Space Agency (AEE) has taken the initiative to strengthen the country’s capabilities in prevention, detection, and response. The objective is clear: to turn the challenge into an opportunity to bolster prevention, detection, and response capabilities, ensuring that Spain has the most advanced tools to protect lives, infrastructure, and the environment.

In collaboration with the European Space Agency (ESA), through its ESA InCubed programme, AEE is launching a pioneering national call for the development of innovative Earth Observation applications.

Read the full article (in Spanish) on www.aee.gob.es.