Editor’s note: One way to search for habitable – and possibly inhabited – worlds in our solar system and in other star systems is to examine planetary atmospheres for chemical compounds or “biosignatures”. The idea being that these compounds are required for the origin of life or are the byproduct of life’s presence. Or both. In the case, the ability to quantify carbon dioxide, a gas that plays an important trole in Earth’s climate and is part of our world’s carbon cycle. CO2 is something that can also be searched for in exoplanet atmospheres are “technosignatures” i.e. atmospheric pollutants that a technological species might create. This paper describes a feature of Earth’s atmosphere that could conceivably factor into some of these searches or models of how habitable worlds operate.
There is a growing urgency to track greenhouse gasses with the resolution, precision and accuracy needed to support independent verification of CO2 fluxes at local to global scales. The current generation of space-based sensors, however, only provides sparse observations in space and time.
This challenge has fueled interest in the potential use of data from existing missions originally developed for other applications for inferring global greenhouse gas variability.
The Advanced Baseline Imager (ABI) onboard the Geostationary Operational Environmental Satellite (GOES-East), operational since 2017, provides full coverage of much of the western hemisphere at 10-minute intervals from geostationary orbit at 16 wavelengths at an approximately 2km2 spatial resolution.
Here, we leverage this high spatial coverage and temporal revisit to develop a single-pixel, physics-guided neural network to estimate dry-air column CO2 mole fraction (XCO2). The model employs a time series of GOES-East’s 16 spectral bands, ECMWF ERA5 lower tropospheric meteorology, MODIS surface reflectance, solar and satellite viewing geometry, and day of year. Training used collocated GOES-East and OCO-2/OCO-3 observations.
We also present case studies illustrating the use of the model to observe XCO2 enhancements over urban areas and drawdown over agricultural regions.
Overall, while the precision of GOES-East derived XCO2 can never rival that of dedicated instruments, the unprecedented combination of contiguous geographic coverage, 10-minute temporal frequency, and multi-year record offers the potential to observe aspects of atmospheric CO2 variability currently unseen from space.
Aaron Sonabend-W, Sean Campbell, John Platt, Christopher Van Arsdale, Anna M. Michalak
Comments: 23 pages, 7 figures, 1 table
Subjects: Atmospheric and Oceanic Physics (physics.ao-ph); Earth and Planetary Astrophysics (astro-ph.EP); Machine Learning (cs.LG)
MSC classes: Primary 86A10, Secondary 86A22, 68T07, 62P12, 62M10
Cite as: arXiv:2605.23991 [physics.ao-ph] (or arXiv:2605.23991v1 [physics.ao-ph] for this version)
https://doi.org/10.48550/arXiv.2605.23991
Focus to learn more
Submission history
From: Aaron Sonabend
[v1] Sun, 17 May 2026 19:27:17 UTC (7,711 KB)
https://arxiv.org/abs/2605.23991
Astrobiology, Gaia,
