Preprints
https://doi.org/10.5194/egusphere-2022-806
https://doi.org/10.5194/egusphere-2022-806
19 Sep 2022
 | 19 Sep 2022

Use of machine learning to retrieve nitrogen dioxide with hyperspectral imagers in the ultraviolet and blue spectral range

Joanna Joiner, Sergey Marchenko, Zachary Fasnacht, Lok Lamsal, Can Li, Alexander Vasilkov, and Nickolay Krotkov

Abstract. Nitrogen dioxide (NO2) is an important trace-gas pollutant and climate agent whose presence also leads to spectral interference in ocean color retrievals. NO2 column densities have been retrieved with satellite UV-Vis spectrometers such as the Ozone Monitoring Instrument (OMI) and Tropospheric Monitoring Instrument (TROPOMI) that typically have spectral resolutions of the order of 0.5 nm or better and spatial footprints as small as 3.5 km × 5 km. These NO2 observations are used to estimate emissions, monitor pollution trends, and study effects on human health. Here, we investigate whether it is possible to retrieve NO2 amounts with lower spectral resolution hyper-spectral imagers such as the Ocean Color Instrument (OCI) that will fly on the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) satellite set for launch in early 2024. OCI will have a spectral resolution of 5 nm and a spatial resolution of ∼1 km with global coverage in 1–2 days. At this spectral resolution, small scale spectral structure from NO2 absorption is still present. We use real spectra from the OMI to simulate OCI spectra that are in turn used to estimate NO2 slant column densities (SCDs) with an artificial neural network trained on target OMI retrievals. While we obtain good results with no noise added to the OCI simulated spectra, we find that the expected instrumental noise substantially degrades the OCI NO2 retrievals. Nevertheless, the NO2 information from OCI may be of value for ocean color retrievals, as our simulations suggest that it will be of similar or slightly better quality as compared with TROPOMI NO2 data at TROPOMI spatial resolution on a daily basis and will be available simultaneously. OCI retrievals can also be temporally averaged over time-scales of the order months to reduce noise and provide higher spatial resolution maps that may be useful for downscaling information provided by lower spatial resolution instruments such as OMI and TROPOMI, for high resolution emissions estimates, and other applications. In addition, we explore the possibility of using an extended fitting window for NO2 retrievals as compared with traditional approaches. We demonstrate that the use of an extended spectral fitting window can reduce random errors in a current state-of-the-art OMI NO2 SCD product. Machine learning approaches with extended fitting windows, once trained, can also substantially speed up NO2 spectral fitting algorithms as applied to OMI, TROPOMI, and similar instruments that are flying or will soon fly in geostationary orbit.

Journal article(s) based on this preprint

26 Jan 2023
Use of machine learning and principal component analysis to retrieve nitrogen dioxide (NO2) with hyperspectral imagers and reduce noise in spectral fitting
Joanna Joiner, Sergey Marchenko, Zachary Fasnacht, Lok Lamsal, Can Li, Alexander Vasilkov, and Nickolay Krotkov
Atmos. Meas. Tech., 16, 481–500, https://doi.org/10.5194/amt-16-481-2023,https://doi.org/10.5194/amt-16-481-2023, 2023
Short summary

Joanna Joiner et al.

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2022-806', Anonymous Referee #2, 11 Oct 2022
    • AC2: 'Reply on RC1', Joanna Joiner, 23 Nov 2022
  • RC2: 'Comment on egusphere-2022-806', Anonymous Referee #1, 17 Oct 2022
    • AC1: 'Reply on RC2', Joanna Joiner, 22 Nov 2022

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2022-806', Anonymous Referee #2, 11 Oct 2022
    • AC2: 'Reply on RC1', Joanna Joiner, 23 Nov 2022
  • RC2: 'Comment on egusphere-2022-806', Anonymous Referee #1, 17 Oct 2022
    • AC1: 'Reply on RC2', Joanna Joiner, 22 Nov 2022

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload
AR by Joanna Joiner on behalf of the Authors (23 Nov 2022)  Author's response   Author's tracked changes   Manuscript 
ED: Referee Nomination & Report Request started (08 Dec 2022) by Michel Van Roozendael
RR by Anonymous Referee #2 (22 Dec 2022)
ED: Publish as is (23 Dec 2022) by Michel Van Roozendael
AR by Joanna Joiner on behalf of the Authors (04 Jan 2023)  Author's response   Manuscript 

Journal article(s) based on this preprint

26 Jan 2023
Use of machine learning and principal component analysis to retrieve nitrogen dioxide (NO2) with hyperspectral imagers and reduce noise in spectral fitting
Joanna Joiner, Sergey Marchenko, Zachary Fasnacht, Lok Lamsal, Can Li, Alexander Vasilkov, and Nickolay Krotkov
Atmos. Meas. Tech., 16, 481–500, https://doi.org/10.5194/amt-16-481-2023,https://doi.org/10.5194/amt-16-481-2023, 2023
Short summary

Joanna Joiner et al.

Joanna Joiner et al.

Viewed

Total article views: 429 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
291 125 13 429 2 4
  • HTML: 291
  • PDF: 125
  • XML: 13
  • Total: 429
  • BibTeX: 2
  • EndNote: 4
Views and downloads (calculated since 19 Sep 2022)
Cumulative views and downloads (calculated since 19 Sep 2022)

Viewed (geographical distribution)

Total article views: 419 (including HTML, PDF, and XML) Thereof 419 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 25 Mar 2023
Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Short summary
Nitrogen dioxide (NO2) is an important trace gas for both air quality and climate. NO2 affects satellite ocean color products. A new ocean color instrument - OCI (ocean color instrument) - will be launched in 2024 on a NASA satellite. We show that it will be possible to measure NO2 from OCI even though it was not designed for this. The techniques we developed here, based on machine learning, can also be applied to instruments already in space to speed up algorithms and reduce effects of noise.