Preprints
https://doi.org/10.5194/egusphere-2023-101
https://doi.org/10.5194/egusphere-2023-101
27 Jan 2023
 | 27 Jan 2023

Applying machine learning to improve the near-real-time products of the Aura Microwave Limb Sounder

Frank Werner, Nathaniel J. Livesey, Luis F. Millán, William G. Read, Michael J. Schwartz, Paul A. Wagner, William H. Daffer, Alyn Lambert, Sasha N. Tolstoff, and Michelle L. Santee

Abstract. A new algorithm to derive near-real-time (NRT) data products for the Aura Microwave Limb Sounder (MLS) is presented. The old approach was based on a simplified optimal estimation retrieval algorithm (OE-NRT) to reduce computational demands and latency. This manuscript describes the setup, training, and evaluation of a redesigned approach based on artificial neural networks (ANN-NRT), which is trained on > 17 years of MLS radiance observations and composition profile retrievals. Comparisons of joint histograms and performance metrics derived between the two NRT results and the operational MLS products demonstrate a noticeable statistical improvement from ANN-NRT. This new approach results in higher correlation coefficients, as well as lower root-mean-square deviations and biases at almost all retrieval levels compared to OE-NRT. The exceptions are pressure levels with concentrations close to 0 ppbv, where the ANN models tend to underfit and predict zero. Depending on the application, this behavior might be advantageous. While the developed models can take advantage of the extended MLS data record, this study demonstrates that training ANN-NRT on just a single year of MLS observations is sufficient to improve upon OE-NRT. This confirms the potential of applying machine learning to the NRT efforts of other current and future mission concepts.

Frank Werner 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 egusphere-2023-101', Anonymous Referee #1, 16 Feb 2023
  • RC2: 'Comment on egusphere-2023-101', Anonymous Referee #2, 20 Feb 2023
  • RC3: 'Comment on egusphere-2023-101', Anonymous Referee #3, 28 Feb 2023

Frank Werner et al.

Frank Werner et al.

Viewed

Total article views: 264 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
175 75 14 264 4 5
  • HTML: 175
  • PDF: 75
  • XML: 14
  • Total: 264
  • BibTeX: 4
  • EndNote: 5
Views and downloads (calculated since 27 Jan 2023)
Cumulative views and downloads (calculated since 27 Jan 2023)

Viewed (geographical distribution)

Total article views: 260 (including HTML, PDF, and XML) Thereof 260 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 23 Mar 2023
Download
Short summary
The algorithm that produces the near-real-time data products of the Aura Microwave Limb Sounder has been updated. The new algorithm is based on machine learning techniques and yields data products with much improved accuracy. It is shown that the new algorithm outperforms the previous versions even when it is trained on only a few years of satellite observations. This confirms the potential of applying machine learning to the near-real-time efforts of other current and future mission concepts.