How satellites and AI enhance emissions intelligence for more-effective climate action

Key to enabling climate action is access to actionable climate data. Tech-enabled emissions intelligence, like Climate TRACE’s work with satellites and AI, can support governments, policymakers, the private sector, and investors in identifying where they can achieve the biggest emissions reductions in the shortest amount of time. // Banner image: Shutterstock//AvigatorFortuner
How satellites and AI enhance emissions intelligence for more-effective climate action
Like
The Forum Network is a space for experts and thought leaders—from around the world and all parts of society— to discuss and develop solutions now and for the future. It aims to foster the fruitful exchange of expertise and perspectives across fields, and opinions expressed do not necessarily represent the views of the OECD.

With the near-and-present urgency of the climate crisis, policymakers need tools and resources that enable bigger, bolder, swifter emissions reductions. Key to enabling such climate action is access to actionable climate data. After all, you cannot manage what you do not measure.

Across all levels of government—national, state and municipal—there is an urgent need to prioritise policymaking around emissions reductions. Scientists at the IPCC and other experts stress that to meet the goal of the Paris Agreement to limit global temperature rise to 1.5°C, we must cut emissions in half by the end of this decade.

That means that policymakers need to identify where they can achieve the biggest emissions reductions in the shortest amount of time. That starts with identifying precisely where emissions are coming from within their jurisdictions, from across broad sectors down to individual facility sources. They need data that illustrate which policies are working well and which ones aren’t over time.

Five challenges with current emissions monitoring

Of course, we know generally where greenhouse gas (GHG) emissions come from—burning coal, natural gas and petroleum; methane emitted by cows and landfills—but many governments have found it challenging to develop a clear accounting of the biggest sources in their regions.

As of November 2022, no nation had submitted a complete accounting of its emissions for 2021 to the United Nations Framework Convention on Climate Change (UNFCCC). This is particularly challenging for Global South countries, as they often have limited resources and systems in place for data collection. At the subnational level—states, provinces and cities—even fewer governments have any baseline data on which to base policies.


Source: UNFCCC Submission Portal

Empowering all countries and regions with actionable emissions data is crucial, but requires overcoming at least five common challenges.

  • High costs. Creating a GHG emissions inventory via a traditional bottom-up approach (e.g. collecting self-reported data from companies and facilities that produce emissions or installing monitoring equipment on emitting facilities) is expensive and time-consuming.

  • Outdated data. Because bottom-up emissions inventories are time-intensive to compile, they often have lag times between the data collection and data publication. A two-year lag is common, and 52 countries have not submitted any emissions inventories covering the past decade.

  • Incomplete geographic coverage. Many otherwise high-quality inventories are not available globally. As a general rule, due to the resources needed to create these inventories, data are often relatively complete in wealthier countries but frequently have significant gaps in the Global South, leaving an incomplete picture of total global GHG emissions.

  • Incomplete sectoral coverage. Most inventories also tend to focus on specific sectors like energy, leaving gaps in harder-to-track sectors such as agriculture and unintentional (“fugitive”) methane emissions. This sectoral incompleteness risks leaving major emissions sources “off the books” of GHG accounting, resulting in policymaking blind spots and missed opportunities for decarbonisation.

  • Lack of granularity. Even when countries and regions do have inventories, they are aggregated to such a high level (e.g. all energy industries or all metal manufacturing industries) that the information is not actionable. Almost no inventories are granular enough to identify individual facilities.

The net effect of these five challenges are data gaps that make decarbonisation more challenging.

Climate TRACE: Using satellites and AI to enhance emissions intelligence

Satellite imagery and other remote sensing technologies can help fill this data gap, as they are available globally and affordably. When combined with artificial intelligence (AI), machine learning (ML) and Big Data approaches, they can provide timely, granular, trusted data for policymakers. Tapping into this independent, technology-centric approach is at the heart of Climate TRACE, a global nonprofit coalition created to make meaningful climate action faster and easier by providing timely and independent data on GHG emissions.

The system uses AI that has been “trained” to spot indicators of GHG emissions in satellite imagery, such as steam plumes from power plants or hot spots in steel plants. These indicators are then linked to ground truth data—collected from physical sensors, government datasets and other sources from data-rich regions—to estimate actual GHG emissions. Once trained, these AI models can be deployed globally, even in regions that have not traditionally had access to GHG emissions data.

The Climate TRACE inventory includes annual emissions across more than two dozen sectors for every country and more than 40 territories. In addition, the inventory represents the most detailed facility-level global inventory of GHG emissions ever compiled, covering data for over 72,000 individual sources worldwide. They represent the top known sources of emissions globally in each sector, including power, oil and gas production and refining, shipping, aviation, mining, waste, agriculture, road transportation and heavy industry.


Climate TRACE map of GHG-emitting facilities across different sectors and geographies

Three key use cases

In this unfolding new era of satellite- and AI-based emissions intelligence, there are numerous use cases for this data. Three examples include:

  • Enabling local action. Provinces/states and city governments have the power to implement sector-wide policy change within their jurisdictions, with large-scale emissions-reduction impacts. In fact, if states and regions from the 10 countries with the largest annual GHG emissions fully implemented all of their existing climate pledges, they could mitigate an additional 16 gigatons of CO2 equivalent per year below current national policies’ emissions projections for 2030. This would also lead to total emission levels close to the range for a 2°C emissions pathway. Climate TRACE has partnered with six regional governments to pilot the development of subnational emissions inventories, and in doing so, has enabled new insights for regional policymakers.

Urban road transportation emissions in the state of Western Cape, South Africa
  • Supply chain decarbonisation. There is increasing pressure on companies to decarbonise their supply chains to address Scope 3 emissions (those not produced directly by reporting organisation, but that the organisation indirectly affects in its value chain). However, due to the opacity of these transactions and the complexity of global supply chains, many companies are unable to quantify emissions from key sectors such as land use change, shipping and mining—particularly if the emissions occur in Global South countries. By using data on emissions from individual facilities, businesses can ensure they are sourcing materials from the lowest-emitting suppliers.

Global steel facilities ordered by emissions intensity can be used to track and reduce supply chain emissions by companies when purchasing steel.
  • Sustainable/climate finance. Environmental, social and governance (ESG) data are critical to mobilising funds for climate action, particularly from the private sector. Reliable, third-party data on whether portfolio companies are meeting emissions-reduction targets can help significantly build trust in such market instruments. This is particularly important in developing countries, which are facing the biggest climate finance gap. One of the major issues with ESG data is that company-level data are only available for purchase from data providers, and represent a significant cost to investors that could otherwise be funding more mitigation action. In addition, many of the datasets currently available are overly reliant on voluntarily, self-reported data, resulting in incomplete results that do not cover all parts of the world. Independent, facility-level data that are freely available in all parts of the world could be a gamechanger for climate finance.

Where we go from here: Putting data into action

Starting in 2024, all parties to the UNFCCC—regardless of whether they are in the Global North or South—will have to submit GHG inventories as part of the Biennial Transparency Reports, which will have much more stringent reporting than currently required. In addition to building their own reporting capabilities, governments are seeking additional data sources to help fill gaps and help with verification/validation.

Tech-enabled emissions intelligence, like Climate TRACE’s work with satellites and AI, is a new tool for governments, policymakers, the private sector, investors and anyone interested in taking or advocating for climate action.





  The OECD is working with Member and Partner countries as well as other international and regional organisations, to identify and evaluate economically efficient and socially responsible policy pathways to achieve net-zero emissions at the global, national and city/regional level. Check out the OECD's work on climate action to learn more!

And read more on the Forum Network: CReDo, The Climate Resilience Demonstrator Project: Collaboration and resilience through connected digital twins, by Sarah Hayes, CReDo Strategic Engagement Lead, Connected Places Catapult

Our infrastructure systems were not designed with climate change in mind. To help achieve more resilient services, CReDo uses data across energy, water and communications assets to build a model of an existing infrastructure system and visualise the interdependencies between the networks. 

Please sign in

If you are a registered user on The OECD Forum Network, please sign in

Go to the profile of André VIEIRA
about 1 year ago

Thank you for this objective and clearly written topic. I am under the impression that measuring GHG is so unreliable that I do not quite understand how the targets to reduce GHG by the end of the decade were set. Climate TRACE seems to be quite a USA centered coalition, it would be nice to see a truly global organization with scientific and knowledge sharing that binds the countries to a certain level of commitment.