MethaneSAT’s mission is to measure total methane emissions from oil & gas operations around the world with the frequency, scale and granularity required to assess both basin- and company-level emissions performance in order to motivate and enable faster, more effective emission reductions. 

With a wide field of view and high precision measurement capabilities, MethaneSAT was built specifically to identify where methane originated, how much was being emitted, and how those emissions changed over time. The data provide a complete picture of total emissions from both widely dispersed and highly concentrated sources across areas both large and small. 

MethaneSAT is a subsidiary of the non-profit Environmental Defense Fund. Until it stopped communicating in June 2025, the MethaneSAT instrument measured methane in target areas across the globe. In addition to oil and gas production, the satellite also collected emissions from the agriculture sector. 

By making this comprehensive, foundational data free and publicly available, MethaneSAT was created to provide transparent, empirical methane emissions data for operators, regulators, and other stakeholders.

While operational, MethaneSAT orbited the Earth 15 times a day, observing target areas approximately 200 km x 200 km. It was built to detect methane emissions as low as 2 parts per billion across a 4 km x 4 km area. 

From its observations, MethaneSAT quantified methane emissions at three different scales: 

• Aggregate methane emissions from smaller, dispersed sources, at rates from about 10 kg/hr/km2. When taken as a whole, these sources can make up a significant majority of methane emissions from an oil and gas basin. 
• Methane emissions rates from entire regions, countries, or oilfields. 
• Methane emissions from high-emitting point sources, at rates from about 500 kg/hour. 

The web portal also includes data collected by MethaneAIR aircraft between 2021 and 2023.

Like other passive remote sensing satellite systems, MethaneSAT did not make nighttime measurements. Cloud cover and variable weather conditions also limited detection capabilities. 

See the following papers to learn more about how the team’s research improved cloud and shadow segmentation performance for MethaneSAT: 

• Spectral Channel Attention Network: A Method for Hyperspectral Semantic Segmentation of Cloud and Shadows (Papers Track)
• Deep Learning for Clouds and Cloud Shadow Segmentation in Methane Satellite and Airborne Imaging Spectroscopy

Methane measurements over water are more technically challenging because passive satellites rely on reflected sunlight, and water surfaces behave differently from land. Our team is developing the capability to measure methane emissions over water, called glint mode. This will enable us to process data on emissions from offshore oil and gas production. 

In oil and gas basins with many operators, attributing emissions to specific operators, facilities, or processes is challenging. For example, if multiple operators manage facilities within a 4 km x 4 km area, attributing emissions from that area to a single operator may not be possible.

MethaneSAT was a critical complement to, not a substitute for, other satellites. 

Other satellites can provide high-resolution data for specific, pre-targeted sites. These are particularly helpful for oil and gas operators who are trying to identify larger leaks throughout their operation or monitor areas prone to large emission events. However, those satellites do not capture the vast number of smaller but pervasive leaks throughout the oil and gas production process. 

Without quantifying these smaller, dispersed emissions – which can account for more than 80% of total emissions in many regions – it is not possible to understand or accurately track total emissions for a basin, region, country, or the world.

MethaneSAT was the first satellite to be conceived and built by a nonprofit environmental organization. The mission includes a global team of experts in remote sensing, astrophysics, methane detection, and computational modeling. 

The partners that helped build, launch, and manage MethaneSAT operations include: 

• BAE Systems 
• Blue Canyon Technologies 
• Environmental Defense Fund 
• Google 
• Scientists at the Harvard School of Engineering and Applied Sciences and Harvard Smithsonian Center for Astrophysics 
• IO Aerospace 
• New Zealand Space Agency 
• Rocket Lab 
• SpaceX

MethaneSAT data can be accessed in the following ways: 

• Read data insights on the MethaneSAT website. These highlights and data stories are published periodically and include curated insights such as comparisons of methane emission loss rates between oil and gas basins. 
• Explore and interact with data on the MethaneSAT web portal. MethaneSAT’s online portal is easily accessible and free for everyone. Users can see where emissions are coming from, how much is being emitted across a wide scale, and how those emissions change over time. Where available, the portal also shows publicly reported oil and gas infrastructure, such as wells and pipelines, and owners of the facilities. 
• Analyze data on Google Earth Engine (GEE). MethaneSAT data are available for advanced analysis on GEE, a Google-owned platform that enables users to visualize and analyze geospatial data. The platform allows the flexibility and power for technical users, such as enabling users to layer MethaneSAT data with other datasets. To get access to MethaneSAT data on GEE, go to Google Earth Engine’s MethaneSAT page and fill out the Request Form under “Terms of Use.” 
  • Commercial entities need a paid subscription to use GEE (pricing determined by Google). 
  • Researchers, nonprofits, academic institutions, and other noncommercial entities can access the GEE platform for free (terms as determined by Google). 
  • GEE users can also ingest the data onto their own cloud platforms via the GEE API. Google egress fees apply. 
• Download data from Google Cloud Platform. Users who prefer not to access the data on Google Earth Engine or ingest via GEE API have the option to download MethaneSAT data directly from Google Cloud Platform. Google egress fees apply. To get access, go to Google Earth Engine’s MethaneSAT page and fill out the Request Form under “Terms of Use.”

The MethaneSAT web portal includes the location, date, time, and methane emission rates at three separate scales: 

• Total emissions per ~200 km x 200 km target area. 
• Dispersed emissions, represented on a gridded heat map (with grid cell sizes such as 4 km x 4 km), which show in more detail where emissions likely originated. 
• Distinct point sources, or methane emissions at very high levels (from around 500 kg/hour) that can be traced to a more specific location. 

The MethaneSAT web portal also includes information from an Oil and Gas Infrastructure Mapping (OGIM) database. This global, spatially explicit, and granular database of oil and gas infrastructure is developed by Environmental Defense Fund and based on public-domain datasets reported by government, industry, academic, and other entities. Visit this page for more information about how this data is collected, managed, and updated.

Distinct point sources are the origin points of plumes that can be traced to a specific location. MethaneSAT was built to detect emissions of approximately 500 kg of methane per hour or more from distinct point sources. MethaneAIR aircraft technology could detect point sources emitting about 200 kg of methane per hour or more. Detected emissions may be from any type of source or any industry.

To learn more about the methodologies EDF used to quantify emissions from point sources detected during our 2023 MethaneAIR campaign, see this paper

Large, distinct plumes of methane can be traced to a specific source, yet our instruments detect a significant amount of additional methane in the atmosphere. 

Dispersed area emissions are our estimates of all detected methane, mapped onto grid cells with sizes such as 4 km x 4 km and then visualized on a heatmap. These emissions can come from many types of sources, including leaky equipment, abandoned infrastructure, and intentional venting. Research shows that these sources, which can easily number in the thousands depending on the geography, can be responsible for up to 80% of total emissions from a given region. 

Identifying these emissions matters because they account for a significant amount of methane – added up, it’s often more methane than is emitted by large, identifiable sources. 

Please see the next question for how to interpret these dispersed area emissions.

Area heatmaps presenting emissions data are subject to uncertainty. A heatmap from a single date is best considered alongside other evidence. Where possible, it is best to look at heatmaps from multiple dates to identify emissions patterns, or compare a heatmap with other data from the area, such as oil and gas infrastructure, or emissions data from other measurement technologies.&

We regularly update the algorithms that calculate emission rates to increase accuracy and scientific confidence: 

• Versions 0.5.0 to 0.7.10 average 4,000 maps to find the best estimate of emissions from each grid cell. This method is called CORE (Constrained and Optimized Retrieval of Emissions), and it accounts for factors such as topography and weather.
• Version 0.9.x improves how the model accounts for areas with very low emissions – so low that the emissions can’t be distinguished from background levels. These grid cells are now flagged with the description “below noise threshold.”

For more about uncertainties and the model versions, see our technical explanation [PDF].

The portal indicates publicly-reported operators and facility types that are located within a 500-meter radius of detected point source emissions. However, the source of those emissions may be a facility that was not publicly reported as of February 2025, or a different type of source.

Even with those limitations, distinct point sources may still be attributable to specific operators or facilities. 

Dispersed emissions may also be attributable, especially in areas where there is a single operator or a small number of operators that own assets across large geographies. 

In complex regions like the Permian Basin, where facilities are operated by different companies in close proximity to each other, direct attribution is more difficult.