Preprints
https://doi.org/10.5194/amt-2021-145
https://doi.org/10.5194/amt-2021-145

  19 Jul 2021

19 Jul 2021

Review status: a revised version of this preprint is currently under review for the journal AMT.

Assessing the Feasibility of Using a Neural Network to Filter OCO-2 Retrievals at Northern High Latitudes

Joseph Mendonca1, Ray Nassar1, Christopher O'Dell2, Rigel Kivi3, Isamu Morino4, Justus Notholt5, Christof Petri5, Kimberly Strong6, and Debra Wunch6 Joseph Mendonca et al.
  • 1Environment and Climate Change Canada, Toronto, ON, Canada
  • 2Cooperative Institute for Research in the Atmosphere, Colorado State University, Fort Collins, CO, USA
  • 3Finnish Meteorological Institute, Sodankylä, Finland
  • 4National Institute for Environmental Studies (NIES), Tsukuba, Japan
  • 5Institute of Environmental Physics, University of Bremen, Bremen, Germany
  • 6Department of Physics, University of Toronto, Toronto, ON, Canada

Abstract. Satellite retrievals of XCO2 at northern high latitudes currently have sparser coverage and lower data quality than most other regions of the world. We use a neural network (NN) to filter OCO-2 B10 bias-corrected XCO2 retrievals and compare the quality of the filtered data to the quality of the data filtered with the standard B10 quality control filter. To assess the performance of the NN filter, we use Total Carbon Column Observing Network (TCCON) data at selected northern high latitude sites as a truth proxy. We found that the NN filter decreases the overall bias by 0.25 ppm (~50 %), improves the precision by 0.18 ppm (~12 %), and increases the throughput by 16 % at these sites when compared to the standard B10 quality control filter. Most of the increased throughput was due to an increase in throughput during the spring, fall, and winter seasons. There was a decrease in throughput during the summer, but as a result the bias and precision were improved during the summer months. The main drawback of using the NN filter is that it lets through fewer retrievals at the highest latitude Arctic TCCON sites compared to the B10 quality control filter, but the lower throughput improves the bias and precision.

Joseph Mendonca 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-2021-145', Anonymous Referee #1, 09 Aug 2021
    • AC1: 'Reply on RC1', Joseph Mendonca, 06 Oct 2021
  • RC2: 'Comment on amt-2021-145', François-Marie Bréon, 09 Sep 2021
    • AC2: 'Reply on RC2', Joseph Mendonca, 06 Oct 2021

Joseph Mendonca et al.

Joseph Mendonca et al.

Viewed

Total article views: 373 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
267 91 15 373 17 4 1
  • HTML: 267
  • PDF: 91
  • XML: 15
  • Total: 373
  • Supplement: 17
  • BibTeX: 4
  • EndNote: 1
Views and downloads (calculated since 19 Jul 2021)
Cumulative views and downloads (calculated since 19 Jul 2021)

Viewed (geographical distribution)

Total article views: 352 (including HTML, PDF, and XML) Thereof 352 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 21 Oct 2021
Download
Short summary
Machine learning has become an important tool for pattern recognition in many applications. In this study, we used a neural network to improve the data quality of OCO-2 measurements made at northern high latitudes. The neural network was trained and used as a binary classifier to filter out bad OCO-2 measurements in order to increase the accuracy and precision of OCO-2 XCO2 measurements in the Boreal and Arctic regions.