ESA title

Category: Philab

Φ-lab’s continuing EO innovation drives AI-derived methane detection

Posted on by nchimonides

ESA Φ-lab, an established prime mover for disruptive innovation in European Earth observation (EO), is supporting a project on identifying methane pipeline leakage using satellite data. The research moves the state of the art forward through a data-fusion approach, employing machine learning algorithms to combine various remote sensing image and other time-series data sources.

Φ-lab has a proven track record for driving innovation in the Earth observation sector, including through the financial and technical sponsorship of cutting-edge research via PhDs.

Methane detection from space is one such area of Φ-lab’s research. Albeit with a shorter lifespan in the atmosphere, methane gives rise to up to 30 times more heat retention than carbon dioxide and so is proving to be a significant contributor to global warming. Detecting and fixing methane leaks is therefore a priority in terms of climate-change mitigation, but in the case of gas pipelines, management of leaks can be hampered by the vast expanses of remote locations that the networks of pipes often traverse.

In recent years Φ-lab has overseen two projects, STARCOP and LEO-GEO4GHG, both of which explored how methane plumes could be detected from satellites using onboard AI. In new research that Φ-lab is jointly supervising with German aerospace company OHB Digital Connect and the Technical University of Munich, PhD student and OHB employee Enno Tiemann is exploiting machine learning to analyse a combination of data sources for plume detection and quantification.

“Currently, methane detection often relies on experts who manually inspect the EO data and create a map of the plume that is then used to determine the emission rate – sometimes with a high degree of uncertainty,” Enno explains. “Whereas previous work has adopted a cascading approach, whereby initial AI detection of large plumes is followed by human analysis of higher-resolution optical or infrared data, our research involves algorithms that simultaneously interrogate multiple satellite datasets to improve detection and reduce uncertainty.”

Although Sentinel-5P is a well-established methane detector, giving daily coverage of the entire planet, it has a fairly coarse resolution that means the minimum measurable emission rate is around 25 tonnes per hour. Sentinel-2 and Landsat 8 and 9 by contrast have a threshold of 1.5-1.8 tonnes per hour, but lack daily coverage. By fusing time series data from multiple satellites, in conjunction with wind-vector data, the new research is expected to yield more robust estimates of plume location and gas flows, together with insight on the evolution of plumes over time.

Φ-lab provides essential support to the project through technical supervision, expertise sharing and joint funding with OHB. Φ-lab AI Applications Lead Rochelle Schneider is the project supervisor for ESA: “In addition to giving vital input to the project, Φ-lab is able to grant external researchers access to our pool of world-leading AI experts and data scientists. We’re very happy to help guide this research initiative, which like STARCOP and LEO-GEO4GHG is an example of AI-driven EO that is sure to make a significant contribution to combatting methane emissions in the energy sector.”

To know more: Φ-lab, OHB, TUM

Photo courtesy of Robzor

Φsat-2 gets two new AI apps

Posted on by nchimonides

Φsat-2, ESA’s groundbreaking cubesat scheduled for launch in June, will now include two new AI-driven apps destined to demonstrate a crucial role in future environmental monitoring from space. The apps, which focus on marine pollution and wildfires, were the winning entries in ESA’s OrbitalAI Challenge.

AI is already established as a vital tool in Earth observation, helping to sift through terabytes of satellite data to provide useful insight for scientists, policy makers and commercial operators. But while most AI processing of information is carried out on the ground once the data have been downloaded, there are significant advantages to be had from integrating AI capabilities on the satellite.

Read the full article on www.esa.int.

Building ChatGPT-style tools with Earth observation

Posted on by nchimonides

Imagine being able to ask a chatbot, “Can you make me an extremely accurate classification map of crop cultivation in Kenya?” or “Are buildings subsiding in my street?” And imagine that the information that comes back is scientifically sound and based on verified Earth observation data. ESA, in conjunction with technology partners, is working to make such a tool a reality by developing AI applications that will revolutionise information retrieval in Earth observation.

A digital helping hand for data

Earth observation generates vast volumes of vital data every day, but it is difficult for humans alone to ensure that we obtain the best value from that data. Fortunately, AI helps in interacting with such large and complex datasets, identifying key features and presenting the information in a user-friendly format.

I*STAR for example, an activity co-funded by the ESA InCubed programme, developed a platform that uses AI to monitor current events like earthquakes or volcano eruptions so that satellite operators can automatically plan the next data acquisitions for customers. The SaferPlaces AI tool, again supported by InCubed, creates flood maps for disaster response teams by merging in situ measurements with satellite data. SaferPlaces was crucial to damage assessment efforts during last year’s floods in Emilia-Romagna in Italy.

Read the full article on www.esa.int.

Hello Major TOM: ESA Φ-lab releases largest ML-ready Sentinel-2 dataset ever published

Posted on by nchimonides

ESA Φ-lab has launched Major TOM (Terrestrial Observation Metaset), a community-oriented project that allows researchers to share, use and combine large Earth observation (EO) datasets. The Major TOM framework will help unlock the huge potential of satellite imagery by offering users the largest ever quality-controlled and globally distributed sample of data, with future expansions to multiple satellites and modalities planned from both Φ-lab and the wider community.

Recent years have seen a marked trend towards larger, more general EO and geospatial models, known as foundation models, which require massive volumes of high-quality training data. These large models present unique opportunities in that they have the potential to help solve many pressing scientific and societal problems.

But there are also challenges, including the risk of deepening the reproducibility crisis seen in AI research, whereby published models are often difficult to recreate due to closed data sources and opaque technical details. Bias is another issue, since all models are skewed by the data they learn from, and this may lead to biases being embedded into the systems that foundation models form part of.

ESA Φ-lab believes that these issues can be alleviated through the creation of high-quality globally distributed and collaborative ML-ready datasets, and has begun to integrate them under the moniker Major TOM. These ML-ready datasets are a means to steer the development of large models in a positive direction, democratising them and helping to make systems that are more reproducible and with a lower bias by virtue of the dataset’s global sampling. To achieve this, Φ-lab has partnered with Hugging Face to host and freely distribute Major TOM on the Hugging Face Hub. With its open and community-driven platform for datasets and models, Hugging Face is a leading light for the democratisation of machine learning technology.

The creation of such a large dataset presented the team with several technical hurdles. “Satellite data is often held and delivered in very large products – over 100 km across – which many people find difficult to work with for machine learning applications, especially when trying to combine different satellites whose products overlap to differing extents,” explains Φ-lab research fellow Alistair Francis. “By contrast, Major TOM uses a fixed, 10 km grid across the entire globe, meaning that data from one Major TOM dataset will fit neatly on top of another.”

Whilst the sheer volume of data processing involved was a challenge, the need to ensure its quality was equally difficult. For example, optical satellite imagery often contains clouds that hide the surface below. Although not eliminated from the Major TOM dataset entirely, cloudy imagery was minimised by using Φ-lab’s state-of-the-art AI cloud mask, soon to be released publicly.

Illustration of the global coverage of Major TOM Core. Regions in colour denote sampled areas (green for land and light blue for sea).

Major TOM’s inaugural core dataset has now been released on Hugging Face. It constitutes the largest ML-ready collection of Copernicus Sentinel-2 images ever published. Covering over 50% of the Earth’s surface (including almost all dry land) with nearly 50 TB of data and 2.5 trillion pixels, Major TOM Core is a game-changer for those seeking to train large models with satellite data. It is expected that future expansions from the broader EO community, spearheaded by ESA Φ-lab, will spawn a diverse ecosystem of combinable datasets that will be invaluable in creating the next generation of large deep learning models from satellite data.

Giuseppe Borghi is the Head of ESA Φ-lab: “We want to build an open community of contributors and end users who can create a data landscape that ensures EO derives the largest benefits from the AI revolution. If we want to make sure that EO models are reliable, reproducible, traceable and in turn, trustworthy, then it stands to reason that we need to start with high-quality trustable data.”

An interview on Major TOM with Alistair Francis and fellow Φ-lab researcher Mikolaj Czerkawski can be found here.

To know more: Φ-lab, Hugging Face, Major TOM paper preprint

Header image contains modified Copernicus Sentinel data (2022), processed by ESA

FOREST-2 to deliver thermal-sensing insights to Copernicus

Posted on by nchimonides

ESA is working with European New Space company OroraTech to demonstrate how data from its temperature-sensing FOREST-2 mission will facilitate the aims of the Copernicus programme.

The Munich-based thermal intelligence specialist was one of nine firms selected as European Emerging Copernicus Contributing Missions (CCMs) in June 2023, following a recruitment drive designed to encourage – and capitalise on – New Space growth in Earth observation. In complement to the Sentinel family, this group of providers will soon supply commercial data to Copernicus to help address key environmental and societal challenges impacting European citizens.

OroraTech was first supported by incubation programmes ESA BIC Bavaria and ESA Kick-Start, and in 2022 was awarded funding from ESA InCubed for the development of its upcoming FOREST-3 CubeSat – all of which served as important stepping stones to the firm joining Copernicus.

Read the full article on spacedata.copernicus.eu.

Image courtesy of OroraTech

ESA Φ-lab broadens its cloud-service support to start-ups

Posted on by nchimonides

As part of its extensive efforts to nurture innovation in Earth observation technologies and applications, ESA Φ-lab is providing its supported companies with preferential rates agreed with two leading cloud-service suppliers in collaboration with the ESA Commercialisation, Industry and Competitiveness Directorate. Scaleway will provide start-ups with a cloud storage and business services package, while Ellipsis Drive is offering deals on its cloud spatial-data management platform.

ESA Φ-lab is a major driving force behind innovation and commercialisation in European Earth observation, lending support to businesses of all shapes and sizes through the ESA InCubed programme and various research initiatives. Help comes not only in the form of funding and technical and commercial support, but also in the creation of purpose-driven partnerships that deliver essential business and commercialisation services. Two such arrangements, drawn up in collaboration with the ESA Partnership Initiative for Commercialisation (EPIC), have now been agreed with premier cloud-service providers Scaleway and Ellipsis Drive.

French company Scaleway supplies cloud infrastructure and services to over 25 000 customers, including more than 700 European start-ups. CEO Damien Lucas explains the nature of Scaleway’s offer within the ESA partnership agreement: “Qualifying businesses from ESA’s portfolio will be fast-tracked through the selection process of our Startup Programme. Upon acceptance, they will have access to a wide range of perks, including cloud credits to be spent on our public cloud products, dedicated expert advice and access to our global community.

“We’re very proud to play our part in helping European space-sector entrepreneurs establish their business systems and we look forward to working with the commercialisation teams in ESA on joint promotional activities.”

Ellipsis Drive provides a cloud-based B2B tool for ingesting, organising and accessing spatial data, with web-based visualisations and integrated use via a myriad of plug-ins, packages and applications. Based in the Netherlands, Ellipsis Drive has hosted data for over 400 customers to date and currently has around 2500 users managing and consuming spatial content from its platform.

“We see partnering with ESA as a clear win-win scenario, giving start-ups access to our spatial data management, visualisation and integration service on very favourable terms, while also enabling us to tap into a broader customer community in the space sector,” says Ellipsis Drive’s Rosalie van der Maas. “We’re offering up to 100GB of free storage for the first year, with substantial discounts for larger plans and subsequent periods.”

Further details on these offers and how to take advantage of them will be communicated to InCubed and other ESA Φ-lab-supported companies in the near future, and will also be distributed via the ESA BIC and ESA Technology Broker Network.

Michele Castorina is InCubed Programme Manager and head of the ESA Φ-lab Invest Office: “The fact that ESA creates partnerships for the benefit of European start-ups is testimony to the across-the-board helping hand that the Agency provides to the space industry. Cloud storage and spatial data hosting are especially relevant for commercial Earth observation, and I fully expect these services to be a significant asset to InCubed-co-funded businesses as they scale up their operations.”

“Having the right tools for their business infrastructure is a vital stepping stone to market success for early-stage enterprises,” added Joana Kamenova, Commercialisation Officer and lead for EPIC at ESA. “This type of collaboration helps to advance the growth of the European space ecosystem, and we will actively promote Scaleway’s and Ellipsis Drive’s offer packages to our networks.”

To know more: ESA Φ-lab, InCubed, EPIC, ESA BICs, ESA Technology Brokers, Scaleway, Ellipsis Drive

Photo courtesy of Fauxels

InCubed announces call for proposals for maritime EO applications

Posted on by nchimonides

Our oceans and seas are a substantive source of economic activity, and it is within this context that ESA InCubed has released a thematic call dedicated to the globally significant maritime sector. Open until March of this year, the campaign focuses on developing innovative products and services using Earth observation for areas such as maritime security, transportation, pollution and fisheries.

As part of its remit to co-fund European commercial development across the entire Earth observation (EO) value chain, the ESA InCubed programme periodically issues thematic calls with particular focus areas. Last year saw the launch of a Cultural and Natural Heritage call, while the latest campaign is on the maritime sector.

The world’s oceans are home to a major component of the global economy, not only in terms of trade and transportation, but also energy and food production. Around 90% of international trade occurs  through maritime routes, and some 50 000 merchant vessels are currently in operation across the globe. Europe’s blue economy is worth nearly €650 billion per annum, and the continent has some of the world’s largest harbours and boasts a number of major shipping companies. But our oceans are also subject to sizeable societal issues and challenges, including marine plastic pollution, climate change effects, overfishing, security issues and illegal migration.

The InCubed Maritime Call seeks to address these challenges by inviting participants to submit ideas for innovative EO solutions that will help to ensure a green and sustainable maritime sector. The call divides the sector into specific areas of interest:

  • Security and emergency
  • Finance and Insurance
  • Renewable Energy
  • Construction
  • Transportation
  • Pollution monitoring
  • Fisheries and acquaculture

Further details on these areas, along with the global and European maritime landscapes and how the sector can derive enormous benefits from EO services, can be found in two highly valuable new reports commissioned by ESA:

These reports give a wealth of useful information, helping applicants from both within and outside the maritime sector to submit ideas for the call under the seven areas of interest. Responses are welcome from individuals, companies and academia in InCubed participating states.

Idea pitches will be assessed based on criteria such as the level of innovation, the credibility of the business opportunity and the quality and completeness of the proposal.

Details of the call can be found on ideas.esa.int, and the closing date for entries is 29 March 2024.

To know more: ESA InCubed

Photo courtesy of George Desipris

Φ-lab contributes to COP 28 discussions on role of AI4EO in health, climate adaptation and risk management

Posted on by nchimonides

Φ-lab contributed to the recent United Nations COP 28 Climate Change Summit in Dubai, sharing ESA’s experience in tackling climate issues through AI-driven modelling and analysis of Earth observation (EO) data. The team played a major role in sessions on climate adaptation in the areas of health monitoring and risk transfer mechanisms.

COP 28, the 28th meeting of the Conference of the Parties, is the UN climate change forum that took place between 30 November and 13 December in Dubai. With over 70 000 delegates from around 200 countries, COP 28 was a truly worldwide gathering that included the first ever global stocktake on climate change.

ESA has or is developing a number of EO climate-related activities that provide valuable insight to scientists, the private sector and policy makers alike, notably the RECCAP-2 and World Emission projects and the Copernicus Sentinel-5P and Anthropogenic Carbon Dioxide Monitoring missions. Within ESA’s Directorate of Earth Observation, Φ-lab has established a formidable reputation as a centre of excellence for transformative innovation, facilitating groundbreaking EO research through enabling technologies such as machine learning and edge and quantum computing.

One of the objectives of the COP global stocktake is to evaluate our collective progress towards climate adaptation, a theme that was taken up in many sessions at the conference, including two that benefited from Φ-lab’s expertise. Satellite Technologies for Climate Adaptation and Health was organised by the ESA Climate Office and chaired and led by Φ-lab AI Applications Lead Rochelle Schneider, a researcher with extensive experience in health monitoring and prediction in projects such as UNICEF-Φ-lab dengue fever modelling.

Φ-lab’s Rochelle Schneider (left) with fellow participants in the ESA booth at COP 28

Rochelle’s talk began by citing World Health Organisation (WHO) statistics on predicted future climate-related fatalities and current levels of poor air quality. The discussions then went into detail on the role of satellites in providing a fundamental source of environmental hazard and air quality data. When these data products are fed into machine learning models, the result is high-resolution air pollution maps of cities and countries that both highlight areas surpassing WHO guideline limits and enable the assessment of human exposure to such pollution.

In their concluding remarks, Rochelle and fellow speaker Maarten Kappelle of UNEP’s Science Division underlined the importance of data scientists and a broad spectrum of experts joining forces to improve how society combats climate change. In this way these multidisciplinary teams will be able to inform adaptation strategies and directly contribute to public health policies.

Rochelle also took part in the Collaborative Futures: Bridging Gaps with AI and EO in Climate Risk Management and Adaptation session. Here the panellists discussed the potential for AI and data from space to aid adaptation through insurance and financial mechanisms, with particular emphasis on the part that the private sector has to play in providing new solutions for climate risk management. In fact this is a strategic development area for the EO industry, which ESA is addressing and supporting via the InCubed commercialisation programme.

The session included several examples of how AI modelling can help overcome challenges for the risk transfer industry in data gaps and accessibility, leading to the possibility of innovative insurance products to help protect exposed communities.

Both panellists and audience were surprised to hear that AI onboard satellites, rather than being a future development, is already a reality at ESA. Rochelle gave examples of the Agency’s extensive work in this area, including Φ-lab-led projects like RaVAEn, FDL NIO and Cognitive Cloud Computing in Space.

“It’s clear that useful climate monitoring is not possible without the global, near-real-time measurement that Earth observation affords,” Rochelle remarked in summing up her participation. “But as these discussions at COP 28 have shown, when we add AI to the mix, we can create a pipeline that takes raw data, processes it intelligently and creates models of trends, empowering governments, institutions and commercial enterprises in their climate-action decision making.”

To know more: Φ-lab, InCubed, ESA Climate Office, COP 28, UNEP

Photos courtesy of UN Climate Change and Maarten Kappelle

Earth Systems Predictability Forum final report maps out path for long-term human wellbeing

Posted on by nchimonides

With support and guidance from ESA Φ-lab, the Earth Systems Predictability (ESP) Forum earlier this year brought together over 150 top experts to explore how Earth observation (EO) in concert with artificial intelligence (AI) and other emerging technologies can help us make decisions on the future of our planet. The final report, which sets forth a far-reaching vision for ESP and its development, is now available.

ESA and the European Commission’s recent announcement on joint efforts to address climate change highlights once again the crucial importance of EO in advancing our understanding of the Earth’s systems and their impact on global warming. ESA has many climate initiatives already in progress, including the Space for a Green Future Accelerators and the Climate Change Initiative, along with developing comprehensive modelling of the planet through Digital Twin Earth and its pivotal role in DestinE. The ESP Forum adds a significant piece to the jigsaw by seeking to bridge the gap between the knowledge gained through climate change sensing and modelling on the one hand and the required climate action and decision-making on the other.

ESP is described as an emerging discipline that combines elements from EO, AI, simulation, visualisation and decision support. It aims to build on Earth digital twins and provide an accessible, distributed platform that is integrated with everyday life. The forum was held in May of this year, organised by Trillium Technologies, ESA Φ-lab and Oxford University. With its solid experience in groundbreaking innovation in AI, machine learning and digital twinning based on EO assets, Φ-lab was well placed to offer a guiding hand on the proceedings and reporting.

The forum was the result of months of extensive preparation, with the specialist participants engaged in shaping the agenda, the open research questions and the suggested innovation approaches to be discussed and developed during the three-day event.

After releasing provisional conclusions from the event in July, the full ESP Forum report has now been published and can be viewed directly at the foot of the page, or online here.

The report affirms that the foundations of ESP rest on fundamental building blocks such as data from Earth and space, trustworthy and generative AI, AI-based weather and climate models and scaled systems. A detailed deployment vision is set out in the text, drawing on existing work in a variety of disciplines comprising Earth digital twinning, large-scale simulations, finance, environmental, social, and corporate governance (ESG), social return on investment, and decision intelligence.

The final section of the report is an exhaustive ESP roadmap that seeks to provide context for the reader to take action. The roadmap includes technical elements linked to areas such as digital twin Earth visualisation, mobile interface technology, onboard ML, MLOps for ESP, and anthropospheric models and data.

Other key findings include:

  • ESP should be envisioned as a ubiquitous decision-support technology, spanning orders-of-magnitude in scale, from lightweight software modules running on mobile devices, to sophisticated cloud-native Earth digital twins
  • Compelling visualisations and interactive interfaces are essential components of ESP technology, helping to craft narratives around planetary stewardship and support confident mission-critical decisions
  • ESP must be developed from the ground-up in close cooperation with the people who will use the technology to make everyday decisions.

To know more: ESA Φ-lab, Trillium Technologies, Oxford University, ESP Forum

ESA promotes radio frequency monitoring as Spire becomes TPM

Posted on by nchimonides

The global company Spire, which specialises in using continuous global monitoring to track aircraft, ships and weather patterns using a large constellation of CubeSats, is now an ESA Third Party Mission.

As Earth’s population continues to grow, acquiring high quality data to help to predict the movement of the world’s resources is a priority. A specialist in this field, providing radio frequency datasets in near real-time, Spire Global recently announced that it has officially joined ESA’s prestigious Earthnet Third Party Mission (TPM) programme. The data portfolio that Spire provides will include GNSS-RO polarimetric data (PRO) from the ESA InCubed co-funded PROGRES activity.

Read the full article on www.earth.esa.int.

Image courtesy of Spire