Carbon Mapper has assembled a team of world-class science and engineering organizations to achieve its public good mission of methane and CO2 monitoring and mitigation.

Teaming JPL and Planet together created a unique solution: a high sensitivity, moderate resolution payload and agile satellite platforms that can deliver the required precision (detection limit), spatial coverage, and temporal sampling (see Table 1). By persistently tracking and pinpointing point source emissions at individual facilities, Carbon Mapper is designed to complement other observing systems capable of tracking net regional emissions and extremely large point sources. We accomplish this observing strategy by tasking our satellites to acquire data over priority areas based on a combination of prior knowledge of infrastructure locations and follow-up based on “tips” from other satellites designed for wider area monitoring. The target deck figure shown here provides an example of potential priority areas in the US and Canada; there are similar patterns in other parts of the world. This is the same basic tasking approach used by Planet’s constellation of Skysat satellites and is enabled by the moderate field of view provided by JPL’s spectrometer payload design. The system includes other observing modes: glint mode for tracking methane emissions from offshore oil and gas platforms and pushbroom mapping for larger area coverage.

Unlike coarser resolution mapping satellites, each Carbon Mapper satellite is tasked to point at high priority areas characterized by important infrastructure such as oil and gas production fields, pipelines, refineries, power plants, regions with large concentrations of livestock and urban areas where landfills, wastewater treatment plants and natural gas distribution systems are common. The coverage figure shown here illustrates some of the observing modes for a representative orbit track (orange line) by a single satellite – which can be acquired in several minutes. For reference, the green shapes indicate oil and gas producing basins in the western United States. The red rectangle indicates a Carbon Mapper pushbroom observation that is about 1200 km in length (22,000 km2 in area). The white rectangles indicate strip-collect observations where the satellite provides ground motion compensation and much higher sensitivity for priority areas at the expense of area coverage (here limited to about 1200 km2 per rectangle). The satellite can also point to the left and right of the ground track for efficient sampling of nearby target areas. In this example, a single satellite could either do the one pushbroom acquisition or the 5 strip-collect acquisitions but not both. The phase 2 constellation with many satellites will be capable of rapidly covering additional area beyond the single snapshot shown here. Additionally, offshore oil and gas platforms can be observed using glint-mode (not shown here). The Carbon Mapper spacecraft are next generation versions of Planet’s established SkySat series, with the agility, data handling capacity and ability necessary to accommodate the JPL imaging spectrometer payload.
Coverage figure: comparing coverage between Carbon Mapper pushbroom (red) and strip-collect (white) observing modes in selected US gas basins (green). Basemap source: Google Earth
Coverage figure: comparing coverage between Carbon Mapper pushbroom (red) and strip-collect (white) observing modes in selected US gas basins (green). Basemap source: Google Earth
Target deck figure: green boxes indicating potential high priority regions for persistent Carbon Mapper observations in the US and Canada. Basemap source: Google Earth
This spectrometer technology is unique in terms of its high optical throughput (more photons = higher precision), spatial and spectral uniformity, stability and low-noise (a function of optical design and electronics). This translates to a system with a high signal to noise ratio (SNR) that is essential for precise measurements of methane. The new design also has an order of magnitude lower scattered light and much improved spectral efficiency compared to the AVIRIS-NG and GAO spectrometers, thanks to a breakthrough in grating fabrication tolerances. The design variant being baselined for Carbon Mapper actually will have much higher SNR than the AVIRIS-NG/GAO class of instruments used in our ongoing program. These instruments are optically “fast” and have a relatively large (>30 cm aperture). When combined with fine (30 meter) spatial resolution, we predict a methane detection limit of 50-150 kgCH4/hr depending on observing mode, wind-speed and surface brightness. Our experience with airborne point source surveys of multiple regions indicates that Carbon Mapper’s detection limit performance should be sufficient to yield a large fraction of global methane and CO2 point source around the globe which will complement other observing systems focused on wider area emissions.

Sampling frequency is a major design driver for Carbon Mapper. The ability to detect intermittent emissions demands more frequent sampling than conventional surface-based protocols (e.g., quarterly to annually). The Carbon Mapper full constellation is designed to provide daily to weekly sampling depending on target prioritization. This is enabled by Planet’s agile satellites which are able to quickly point at target areas in both the cross- and along-track directions, as well as the spectrometer’s tolerance to relatively large off-nadir angles.

In addition to methane, Carbon Mapper is capable of tracking emissions of the other major carbon molecule, CO2, from large industrial facilities. We have demonstrated the ability to detect and quantify CO2 emissions from in the US and internationally with both our airborne prototypes and experimental satellites (e.g., PRISMA) and predict Carbon Mapper will have a detection limit of about 300,000 kgCO2/hr for a 5 m/s wind speed. This is sufficient to directly image and track CO2 emissions at 90% of the world’s coal power plants along with many refineries and large gas power plants. Most cement and other industrial process emissions are likely below our detection limit.