Derivation of the Carbon Dioxide Total Column in the Atmosphere from Satellite-Based Infrared Fourier-Transform Spectrometer IKFS–2 Measurements: Analysis and Application Experience

Abstract—Based on retrospective comparison with the data of ground-based spectroscopic measurements carried out in Peterhof by St. Petersburg State University (SPbSU) and aircraft measurements carried out in the area of the Novosibirsk Reservoir by the Zuev Institute of Atmospheric Optics in 2019–2022, results of application of a new version of the regression technique for determining the total carbon dioxide XCO2 content (the mole fraction of atmospheric CO₂ in dry air) by measurements of the IKFS-2 infrared Fourier spectrometer of the Meteor-M No. 2 Russian meteorological satellite are analyzed. A description of changes made in the technique to improve the accuracy of satellite estimates is given. For example, to compensate for the influence of changes in IKFS-2 characteristics during a long flight on the XCO2 estimates, they are calibrated based on the results of ground measurements from the NOAA observatory at Mauna Loa volcano (the island of Hawai’i). After calibration and filtering of cloud scenes, the divergence of satellite estimates from ground and aircraft measurements is characterized by a mean square deviation of ~4 ppm or 1% of the total XCO2 content. To speed up the adaptation of the regression algorithm for XCO2 estimation to the IKFS-2 data, it is proposed to use on the new satellites XCO2 estimates from the TCCON ground-based network in addition to the contact measurements of CO₂ concentrations. Also, it is reasonable to use in the regressions the thickness of the cryodeposit on the IKFS-2 photodetector glass as another predictor characterizing the state of the instrument.