Does the Satellite Sensing Column-Averaged Dry-Air Mole Fraction CO₂ (XCO₂) Require Oxygen Concentration From O₂–A Band for Normalizing?

The measured CO<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> mixing ratio is significantly influenced by atmospheric water vapor due to its rapid spatial-temporal variability. In greenhouse gas remote sensing, the dry-air mixing ratio is commonly used to represent CO<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> concentration, which reduces errors arising from water vapor fluctuations. Early satellite spectrometers were often equipped with the O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>-A band to retrieve atmospheric O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> concentration. Leveraging the stable ratio between O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> and dry air, CO<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> can be normalized to derive its dry-air mixing ratio. This approach not only reduces systematic errors but also plays a crucial role in minimizing instrumental biases, thereby enhancing measurement accuracy. However, advancements in satellite instrument calibration have diminished the necessity for O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>-A band normalization. Therefore, this study proposes an alternative normalization method based on meteorological data to alleviate hardware requirements and costs. CO<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> retrieval results from O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>-A band normalization and meteorological data normalization were compared using observations from the Orbiting Carbon Observatory-2 (OCO-2) satellite, and these results were validated against data from the Total Carbon Column Observing Network (TCCON).The findings indicate that, in comparison to TCCON, the mean biases of the O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>-A band scheme are 0.17 ppm, 0.25 ppm, and - 0.005 ppm in glint, nadir, and target modes, whereas the biases for the meteorological normalization method are 0.35 ppm, 0.45 ppm, and 0.19 ppm, respectively. Overall, the systematic bias of the meteorological normalization method is approximately 0.2 ppm greater than that of the O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>-A band scheme, potentially attributable to the spatial-temporal resolution limitations of the Modern-Era Retrospective analysis for Research and Applications, Version-2 (MERRA-2) dataset. Nonetheless, this method remains a viable alternative to O<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub>-A band normalization for CO<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> retrieval.