In addition to tracking methane emissions, Carbon Mapper will also monitor fossil fuel CO2 emissions and deliver 25 other environmental indicators useful for ecosystem management and incident response on land and at sea. This provides significant flexibility that supports commercial services sufficient to build out the fully operational constellation while sustaining Carbon Mapper’s public good mission of delivering transparent methane and CO2 emissions data.

Carbon Mapper will monitor priority regions around the world to detect, pinpoint and quantify methane and CO2 emissions from individual facilities including highly intermittent activity. Our system’s agility and flexible targeting also allows for rapid follow-up for hotspots identified by wider-area mapping satellites such as the European Sentinels and EDF’s MethaneSAT. Carbon Mapper is intended to serve as a major tier of an emerging global system of observing systems and complement the other satellite programs.

Ammonia and carbon dioxide are just some of the gases that can be co-emitted with methane from some landfills, dairies, oil fields and power plants. The data presented here is limited to methane given our research focus on greenhouse gases. Some state and local agencies such as the California Air Resources Board (CARB) and Air Quality Management Districts monitor air quality and criteria pollutants. CARB and the US Environmental Protection Agency also maintain databases of self-reported emissions from large facilities in selected sectors.

A methane source is a condensed surface feature or infrastructure component (typically < 10 meters across) that emits plumes of concentrated methane. In cases where we observe multiple instances of a plume at a given location we combine our individual plume emission estimates to derive a source-average emission rate, including the impacts of intermittency.

The relative amount of methane in a given volume of air is often represented as the methane concentration (or more accurately, the dry air mole fraction). For example, the current global average concentration of near-surface atmospheric methane on earth is about 2 parts per million (ppm).

When methane concentrations in a localized volume of air are measurably higher than average background concentrations (methane in less polluted nearby air) we refer to that increase as the methane enhancement. For example, in the presence of a steady breeze, if a methane emitter results in localized concentrations of 10 ppm and the cleaner upwind air has an average background concentration of 2 ppm we would say that local enhancement is 10−2 ppm = 8 ppm. Methane enhancements can also be represented by absolute mass units; for example, a localized enhancement might contain several kilograms of methane.

An emission is the rate at which methane is generated by a source. For example, the emission rate for a venting gas pipeline compressor station might be 100 kilograms per hour (kg/hr) of methane, equivalent to about 5400 standard cubic feet per hour (scfh) of natural gas.

An emitter (or emission source) refers to the physical or biological activity that generates methane gas. There are two basic categories of methane emitters: point sources and area sources.

We define a point source to be a condensed surface feature or infrastructure component (typically < 10 meters across) that emits plumes of highly concentrated methane that’s above the detection limit of our remote sensing technology. Examples of point sources include individual pieces of natural gas infrastructure, oil wells, refineries, gas-capture systems in landfills, waste water treatment plants, manure management systems at large dairies and wildfires. Point sources include a relatively small number of so-called “super-emitters” with methane emissions >> 10 kg/hour and a large number of much smaller emitters.  The majority of methane plumes in this data portal are super-emitters.

An area source is typically distributed over a large area (typically 1–100 km across) that releases methane in a more diffuse fashion such as anaerobic decomposition occurring with rice cultivation and natural wetlands or enteric fermentation from livestock.

Sometimes the cumulative effect of a large number of low-emitting point sources can manifest as an area source—for example, cities can appear as a large area source due to the combined impact of millions of small natural gas leaks in distribution infrastructure and even downstream of meters in homes and businesses. While our research program includes observations and analysis of methane across multiple scales the data products in this portal are currently primarily limited to local scale methane plumes. 

Depending on the underlying mechanisms, methane emissions can vary by hour, season or even years. For example, methane emissions from oil and gas infrastructure, refineries, power plants and manure management at large dairies are often highly intermittent on time scales of hours due to normal operations such as periodic venting or flaring, but a malfunction or leak can result in emissions that persist for years. In other cases, methane emissions can sometimes vary strongly from season to season as a function of environmental conditions that affect methane producing bacteria (landfills) or by demand (residential natural gas consumption). Over time spans of years, we may see trends in emissions in response to mitigation (reduction) or new activity (growth).

1. High emission point source activity is common and occurs in multiple economic sectors including fossil fuel production and use, waste management and agriculture, with significant regional variations. In many cases, these high emission point sources contribute up to 20-50% of regional total methane emissions.
2. Many methane point sources are highly intermittent while others are persistent –  the average source persistence or frequency is about 25% for most regions.
3. Frequent measurements over large areas can help separate persistent activity (including leaks) from more intermittent activity (including periodic planned maintenance events and a mix of normal and anomalous venting).
4. Sustained, high resolution remote sensing of methane plumes can pinpoint emissions sources and identify specific equipment for efficient follow up by facility operators and regulators.
5. Accurate quantification of regional methane budgets and inventories benefits from tiered observing systems that apply multiple measurements from different vantage points (e.g., land, air and space) to provide a complete picture of the different types of emitters, including strong point sources and wider area sources.

We have research collaborations with facility operators for many locations where we’ve detected methane point sources. In those cases, we shared our methane image data and source coordinates and the operators responded with ground based follow-up measurements to verify and/or further pinpoint the locations. Some of those follow-up efforts exposed malfunctioning hardware or leaks that the operator was able to repair. Examples include natural gas infrastructure in cities such as leaking bypass valves at compressor stations and low-pressure distribution pipelines. In those cases the relevant gas utilities confirmed and repaired the leaks. We have also shared data over the course of several years with operators of large municipal landfills in California who used it to guide improvements in their gas capture systems. Carbon Mapper will continue to share data with operators as the program grows.  

There are two main factors that cause cows and other ruminants to emit methane: enteric fermentation and manure management. Enteric fermentation is a part of the digestive process for cows which causes methane to be expelled from the animal from the front end (burping). On average, roughly half of methane emissions from livestock come from enteric fermentation. The remaining livestock methane emissions are due to manure management.  The majority of methane data in this portal attributed to livestock is primarily associated with a particular class of manure management: anaerobic “lagoons” that store animal waste/manure. Hence, the majority of livestock methane emissions are not detected with these point source measurement methods. 

No, only a subset. The methane point source data includes manure management sources and specifically from “wet management” techniques. Enteric fermentation manifests as an area source best quantified with other methods and is not included in the current version of this web site.

As part of its public good mission to provide precise, timely, accessible, and actionable emissions data, Carbon Mapper makes its data free and open to the public for non-commercial research purposes. Data is available for use within our data portal or by downloading plume and source data directly.

By creating an account and registering for our API, users gain the ability to seamlessly integrate the data into their applications and research projects, in addition to enabling bulk download functionality. Accessing Carbon Mapper data implies acceptance of our Terms of Use. If you have questions about additional licensing options, please don’t hesitate to reach out to us at data@carbonmapper.org.

Our protocols for plume detection and Quality Control (QC) protocols can be found here. Additionally, the following journal papers and reports describe the analysis methods used to generate the various data sets presented here including quantitative emission estimates and uncertainties. Please use the appropriate citation when referencing a report or paper.

Multi-basin analysis: San Joaquin, Permian, Uinta, Denver-Julesburg, Marcellus (Data from 2020-2021)

Cusworth, D. H., Thorpe, A. K., Ayasse, A. K., Stepp, D., Heckler, J., Asner, G. P., Miller, C. E., Chapman, J. W., Eastwood, M. L., Green, R. O., Hmiel, B., Lyon, D., & Duren, R. M. (2022). Strong methane point sources contribute a disproportionate fraction of total emissions across multiple basins in the U.S. PNAS. https://www.pnas.org/doi/10.1073/pnas.2202338119 

Gulf of Mexico offshore platforms (Data from 2021)

Ayasse, A. K., Thorpe, A. K., Cusworth, D. H., Kort, E. A., Negron, A. G., Heckler, J., Asner, G., & Duren, R. M. (2022). Methane remote sensing and emission quantification of offshore shallow water oil and gas platforms in the Gulf of Mexico. Environmental Research Letters, 17(8), 084039. https://doi.org/10.1088/1748-9326/ac8566 

Permian point-source data (Data from 2019)

Cusworth, D. H., Duren, R. M., Thorpe, A. K., Olson-Duvall, W., Heckler, J., Chapman, J. W., Eastwood, M. L., Helmlinger, M. C., Green, R. O., Asner, G. P., Dennison, P. E., & Miller, C. E. (2021). Intermittency of large methane emitters in the Permian Basin. Environmental Science & Technology Letters, 8(7), 567–573. https://doi.org/10.1021/acs.estlett.1c00173

California methane point-source data (Data from 2016-2017) 

Duren, R. M., Thorpe, A. K., Foster, K. T., Rafiq, T., Hopkins, F. M., Yadav, V., Bue, B. D., Thompson, D. R., Conley, S., Colombi, N. K., Frankenberg, C., McCubbin, I. B., Eastwood, M. L., Falk, M., Herner, J. D., Croes, B. E., Green, R. O., & Miller, C. E. (2019). California’s methane super-emitters. Nature, 575(7781), 180–184. https://doi.org/10.1038/s41586-019-1720-3

Duren, R., Thorpe, A., & McCubbin, I. (2020). The California Methane Survey Final Report, CEC-500-2020-047. https://ww2.energy.ca.gov/2020publications/CEC-500-2020-047/CEC-500-2020-047.pdf

When using Carbon Mapper imagery or data products, please provide attribution to Carbon Mapper as follows 

Citing Imagery & Data in Figures (examples)

Methane Imagery © Carbon Mapper https://data.carbonmapper.org 

Methane Imagery and Data © Carbon Mapper, https://data.carbonmapper.org 

You may replace the © symbol with “copyright” or “(c)”. 

You may hyperlink the title Carbon Mapper to https://data.carbonmapper.org in lieu of listing the full URL. 

Citing Data in Papers & Reports

Primary data:

Carbon Mapper data [year(s) of data acquisition]. Retrieved from https://data.carbonmapper.org [day month year of data access]

The use of Carbon Mapper data for non-commercial purposes is subject to Modified Creative Commons Attribution Share Alike 4.0. Please contact Carbon Mapper at data@carbonmapper.org if you would like to request to use our data for any purposes that are not in line with our Terms of Use.

Data on our portal is not real-time due to the complexity of measuring and analyzing methane and CO2 emissions. While some measurement platforms that we use, such as AVIRIS-NG and GAO, have onboard real-time methane detection capabilities, there are several factors that require subsequent data processing and quality control review to validate, pinpoint, and quantify emission sources.  That process requires additional time to complete following each collection. In the future, our goal is to release methane and CO2 emissions data as quickly as thirty days following  observation.  We do offer lower-latency data products (typically available within a few days after each observation) to collaborating research partners. However, those quick-look products are  subject to revision following full processing and review prior to publication.

Pioneering imaging spectrometer sensors are expanding our ability to monitor methane and CO2 emissions on a global scale. While most of our current data set was collected using the AVIRIS-NG and GAO airborne imaging spectrometers, we are now leveraging the power of advanced satellite platforms like EMIT to provide comprehensive and timely insights into emissions patterns worldwide. Our Coalition’s soon-to-be-launched Tanager satellites will offer improved coverage and monitoring capabilities, enabling us to track emissions trends and identify opportunities for mitigation. One important consideration for imaging spectrometer sensors is their minimum detection limits, which is generally understood to be the lowest emission rate that a technology can detect given certain environmental conditions (e.g., wind speed, ground reflectivity) that has been validated with ground-truth data. The minimum detection limits of the sensors we use vary with the platform altitude, ground speed, surface brightness, wind speed and sun angle.  However, generally speaking, the minimum detection limits for AVIRIS-NG and GAO are typically about 10 kg/hr for methane and 10,000 kg/hr for CO2 for a 3 m/s wind and average conditions.  We estimate that EMIT’s minimum detection limit is about 250 kg/hr for methane and 250,000 kg/hr for CO2 for 3 m/s wind and average conditions.

Currently, our portal includes data from airborne field campaigns over the past several years for selected regions, as well as more recent observations around the globe from NASA’s EMIT instrument (which to date has been primarily focused on observing arid mineral dust-emitting regions). The density of methane and CO2 point sources shown in our portal is dependent on three things: 

1.  The inherent prevalence of emission sources within a given region — this varies by emission sector, equipment type and age, relevant regulations, operator practices, and other factors.
2. How much area was observed.
3. How many times and how often the area was observed.

 None of the observing systems we’re currently using were designed for routine global monitoring, but rather have focused on research studies of selected regions.  Some of these regions have been observed many times over several months or years. Other regions have only been observed once or twice so far.  This frequency of observation  is important as many sources are highly variable and/or intermittent. Generally speaking, more observations translates to improved rates of detection and better estimates of average emissions.   

Stay tuned for future expansions in both area coverage and sample frequency as we extend our airborne surveys to other regions, continue to process data from the EMIT sensor, and begin sustained operational monitoring of priority regions around the world with first Carbon Mapper Coalition satellites in 2024.

A source is defined as the specific geographic location from which emissions originate. A point source is a type of source that’s associated with a large emission from a concentrated area, so the geographic location has a specific latitude and longitude.   The emissions from a point source lead to an excess mass or concentration in the atmosphere called a plume. A plume is the atmospheric manifestation of emission processes that occur in a variety of economic sectors. Our portal allows users to view individual methane and CO2 plumes from a variety of sources around the globe. At each source, the user can view individual plume images and instantaneous emission rates associated with each plume as well as estimated average emission rates for the point source based on multiple observations of plumes and/or null detections. To learn more about plumes and sources and how they fit into our overall data product structure, please see our product guide.

The emission rate for any source varies to some degree over time while it is active; we refer to this as variability.  Additionally, many sources are also intermittent — meaning there are periods of time where they stop emitting altogether or drop below sensor detection limits.  The latter is particularly common with industrial activities that operate on some cycle.  To help understand intermittency, we report the persistence of every source, which is simply the number of days with plumes detected divided by the number of days with observations of that source. For example, if a source is observed on 10 days and plumes are detected at the source on 2 of those days then the persistence is 2/10 or 20%.  Note, this method for calculating persistence does not count additional observations and detections within a given day. 

When reporting a source emission rate, we take the average of all emissions at the source multiplied by the estimated persistence. This accounts both for variability in emission rates when a source is active and for periods of inactivity for intermittent sources.  For the example above, if our estimated emissions on day one were 100 kg/hr and on day two were 300 kg/hr then the source emission rate would be (100+300)/2 * 20% = 40 kg/hr.  Since the confidence in our persistence estimates as well as the source emission rate degrades with fewer observations we include a “CAUTION: LOW NUMBER OF SAMPLES” note for sources with less than 4 observations. 

In cases where a source was observed by multiple instrument platforms whose detection limits differ (e.g., AVIRIS-NG and EMIT), persistence is calculated using only the sensor with the lower detection limit. This is done to minimize bias in persistence calculations: instruments with higher detection limits (e.g., EMIT) may fail at detecting lower-emitting persistent sources that may have been detected by other instruments (e.g., AVIRIS-NG, GAO). Expanding on the example above (10 days of observation with 2 point source detections), if EMIT observed on 10 additional days, and detected zero plumes, the average emission rate and persistence would remain unchanged. If instead EMIT observed a single plume at 400 kg/h, the average source emission rate would be updated, but the persistence would remain the same, yielding an average emission rate of (100 + 300 + 400)/3 *20% = 53 kg/h.

Descriptions for fields in our plume and source lists can be found in our product guide, which also provides valuable information about Carbon Mapper products overall. Users can explore a list of key definitions and acronyms, delve into sensor details, get an overview of the data and products, and access detailed information about each product level, all in addition to the field descriptions.

Carbon Mapper plumes are time-tagged with  Coordinated Universal Time, otherwise known as UTC. This is the primary time standard by which the world regulates clocks and time. UTC time can be easily converted to the local time of each plume detection by adding or subtracting the difference between the  local time zone and UTC for a given date. The difference is known as the time offset or UTC offset.

We offer bulk download capabilities to registered users of our data portal. It is generally more efficient to use these bulk download capabilities to access the vast majority of Carbon Mapper Plume data rather than using the API. Bulk downloads are available to registered users by clicking on the icon with the down arrow in the left toolbar of our data portal. These include plume CSVs, GeoTIFFs, and RGBs and are generally current to within the last 3 months.  If you need more recent data, you can use the search filters and selection and download features at the source detail level.

We provide a coverage overlay that can be accessed in the “layers” menu on the right toolbar of the data portal. When this coverage overlay is turned on, it displays the geographic boundaries of each satellite or airborne observation, also known as “scenes.” Users can click on these scenes to view more information, including observation date, sensor platform, and plume identifier (for any plumes that were detected within the scene). This scene information can also be found in our API endpoints and will be downloadable from the data portal in the near future.

 When a sensor observes infrastructure capable of emitting methane or CO2 but no plume is seen, we refer to this as a null detect. This can occur when the source is not emitting at all, or when a source is emitting below the sensor’s detection limit. When it comes to determining average emission rates, accounting for null detects is equally important to tracking observed emissions. Additionally, null detects can provide a powerful tool for verifying that leak repairs are successful (within the detection limit of the sensor).   

In the source list menu, users can check the “Include Null Detects” box to display the results of all observations of a known point source, regardless of whether a plume was detected. This is helpful when viewing the time history of a given source, particularly for sources that are intermittent or to verify that plumes are no longer observed after a leak has been repaired. If a plume is not detected for a given observation, a “no plume detected” label appears in the plume list, and a gray dot representing zero emissions (above the sensor detection limit) appears in the time series plot.

Not necessarily. In many cases, observed methane point sources are expected “process emissions” and to some extent are already accounted for in inventories and reporting systems; for example, venting from vapor relief valves on gas storage tanks, small losses from compressors, etc. Examples of malfunctions include persistent or intermittent leakage due to faulty seals or valves in gas infrastructure, unlit flares, or a leaking gas capture system at a landfill. The degree of intermittency or frequency of an emission source can sometimes  provide a clue about whether a source is an expected process emission or a leak/malfunction.

Carbon Mapper point source detection and quality control processes and protocols require that any detected atmospheric plume must be related to a credible point source on the earth’s surface before reporting. Any observed enhancements that fail to meet QC checks are noted for potential follow-up study but do not result in published plumes or emission rate estimates. In some instances, NASA EMIT sensor detects large methane enhancements that may not necessarily correspond to a large point source nearby. Not all observed atmospheric enhancements are the result of a point source emission, nor can those enhancements always be attributed to a specific emission source. The magnitude of a localized atmospheric enhancement depends on both the emission rate and the degree of local ventilation. Wind speed is usually the biggest factor in ventilation but other atmospheric conditions and local terrain can also play a role, in some cases resulting in “pooling” of methane gas leading to large enhancements near a relatively small emitter. Additionally, point source plumes are a subset of a broader class of CH4 or CO2 enhancements that may occur anywhere in the atmosphere as a result of point source and/or diffuse area sources that may or may not be co-located with the enhancements (e.g., a “cloud” of enhanced CH4 can appear in  the atmosphere some distance downwind of the actual source). 

In some cases, we are confident that a methane or CO2 plume is real, but we may not be prepared to generate a quantitative emission rate estimate.  In such cases, the value is given as “not yet quantified” when the algorithm used to estimate emission rate produced a low-confidence result due to plume shape, surface artifacts and/or issues with wind speed data. Alternative methods used to generate emission rates may be improved over time and released to the public after quality review. 

As part of our plume detection and quality control review process, Carbon Mapper analysts use a combination of high resolution visible images and available infrastructure databases to attribute an observed plume to the most likely emission sector.  We follow sectoral definitions from the Intergovernmental Panel on Climate Change (IPCC). This sectoral attribution is done on a best-effort basis and we welcome feedback on any errors.  Currently, we do  not attribute plumes to equipment type or facility owners/operators because in most cases we lack reliable asset level data.  Asset data is incomplete or not readily available in many regions, and/or in many cases it may be outdated.

Please send an email to data@carbonmapper.org Please note that information submitted to Carbon Mapper may be shared publicly, and we cannot guarantee confidentiality.

If you have specific questions regarding Carbon Mapper data that are not addressed in this FAQ, please email data@carbonmapper.org