27 Oct 2021
27 Oct 2021
Status: this preprint is currently under review for the journal AMT.

On the potential of a neural network-based approach for estimating XCO2 from OCO-2 measurements

François-Marie Bréon, Leslie David, Pierre Chatelanaz, and Frédéric Chevallier François-Marie Bréon et al.
  • Laboratoire des Sciences du Climat et de l’Environnement/IPSL, CEA-CNRS-UVSQ, Université Paris-Saclay, F-91191 Gif-sur-Yvette, France

Abstract. In David et al (2021), we introduced a neural network (NN) approach for estimating the column-averaged dry air mole fraction of CO2 (XCO2) and the surface pressure from the reflected solar spectra acquired by the OCO-2 instrument. The results indicated great potential for the technique as the comparison against both model estimates and independent TCCON measurements showed an accuracy and precision similar or better than that of the operational ACOS (NASA’s Atmospheric CO2 Observations from Space retrievals – ACOS) algorithm. Yet, subsequent analysis showed that the neural network estimate often mimics the training dataset and is unable to retrieve small scale features such as CO2 plumes from industrial sites. Importantly, we found that, with the same inputs as those used to estimate XCO2 and surface pressure, the NN technique is able to estimate latitude and date with unexpected skill, i.e. with an error whose standard deviation is only 7° and 61 days, respectively. The information about the date mainly comes from the weak CO2 band, that is influenced by the well-mixed and increasing concentrations of CO2 in the stratosphere. The availability of such information in the measured spectrum may therefore allow the NN to exploit it rather than the direct CO2 imprint in the spectrum, to estimate XCO2. Thus, our first version of the NN performed well mostly because the XCO2 fields used for the training were remarkably accurate, but it did not bring any added value.

Further to this analysis, we designed a second version of the NN, excluding the weak CO2 band from the input. This new version has a different behaviour as it does retrieve XCO2 enhancements downwind of emission hotspots, i.e. a feature that is not in the training dataset. The comparison against the reference Total Carbon Column Observing Network (TCCON) and the surface-air-sample-driven inversion of the Copernicus Atmosphere Monitoring Service (CAMS) remains very good, as in the first version of the NN. In addition, the difference with the CAMS model (also called innovation in a data assimilation context) for NASA Atmospheric CO2 Observations from Space (ACOS) and the NN estimates are significantly correlated.

These results confirm the potential of the NN approach for an operational processing of satellite observations aiming at the monitoring of CO2 concentrations and fluxes.

François-Marie Bréon et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on amt-2021-313', Christopher O'Dell, 17 Jan 2022
  • RC2: 'Comment on amt-2021-313', Sihe Chen, 22 Jun 2022

François-Marie Bréon et al.

François-Marie Bréon et al.


Total article views: 701 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
568 120 13 701 8 12
  • HTML: 568
  • PDF: 120
  • XML: 13
  • Total: 701
  • BibTeX: 8
  • EndNote: 12
Views and downloads (calculated since 27 Oct 2021)
Cumulative views and downloads (calculated since 27 Oct 2021)

Viewed (geographical distribution)

Total article views: 704 (including HTML, PDF, and XML) Thereof 704 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 02 Jul 2022
Short summary
The estimate of atmospheric CO2 from space measurement is difficult. Current methods are based on a detailed description of the atmospheric radiative transfer. These are affected by significant biases and errors, and are very computer intensive. We have proposed to use instead a neural network approach. A first attempt led to confusing results. Here we provide an interpretation for these results, and describe a new version that leads to high quality estimates.