Preprints
https://doi.org/10.5194/gmd-2021-420
https://doi.org/10.5194/gmd-2021-420
Submitted as: model evaluation paper
01 Feb 2022
Submitted as: model evaluation paper | 01 Feb 2022
Status: a revised version of this preprint was accepted for the journal GMD and is expected to appear here in due course.

A comparative analysis for deep learning model (hyDL-CO v1.0) and Kalman Filter to predict CO in China

Weichao Han1,, Tai-Long He2,, Zhaojun Tang1, Min Wang1, Dylan Jones2, and Zhe Jiang1 Weichao Han et al.
  • 1School of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui, 230026, China
  • 2Department of Physics, University of Toronto, Toronto, ON, M5S 1A7, Canada
  • These authors contributed equally to this work.

Abstract. The applications of novel deep learning techniques in atmospheric science are rising quickly. Here we build a hybrid deep learning (DL) model (hyDL-CO), based on convolutional neural networks (CNN) and long short-term memory (LSTM) neural networks to provide a comparative analysis between DL and Kalman Filter (KF) to predict carbon monoxide (CO) concentrations in China in 2015–2020. We find the performance of DL model is better than KF in the training period (2015–2018): the mean bias and correlation coefficients are 9.6 ppb and 0.98 over E. China, and −12.5 ppb and 0.96 over grids with independent observations. By contrast, the assimilated CO concentrations by KF exhibit comparable correlation coefficients but larger negative biases. Furthermore, DL model demonstrates good temporal extensibility: the mean bias and correlation coefficients are 95.7 ppb and 0.93 over E. China, and 81.0 ppb and 0.91 over grids with independent observations in 2019–2020, while CO observations are not fed into the DL model as an input variable. Despite these advantages, our analysis indicates a noticeable underestimation of CO concentrations at extreme pollution events in the DL model. This work demonstrates the advantages and disadvantages of DL models to predict atmospheric compositions in respective to traditional data assimilation, which is helpful for better applications of this novel technique in future studies.

Weichao Han et al.

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2021-420', Anonymous Referee #1, 28 Feb 2022
    • AC1: 'Reply on RC1', Zhe Jiang, 18 Apr 2022
  • RC2: 'Comment on gmd-2021-420', Anonymous Referee #2, 06 Apr 2022
    • AC2: 'Reply on RC2', Zhe Jiang, 18 Apr 2022

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2021-420', Anonymous Referee #1, 28 Feb 2022
    • AC1: 'Reply on RC1', Zhe Jiang, 18 Apr 2022
  • RC2: 'Comment on gmd-2021-420', Anonymous Referee #2, 06 Apr 2022
    • AC2: 'Reply on RC2', Zhe Jiang, 18 Apr 2022

Weichao Han et al.

Weichao Han et al.

Viewed

Total article views: 474 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
375 82 17 474 6 6
  • HTML: 375
  • PDF: 82
  • XML: 17
  • Total: 474
  • BibTeX: 6
  • EndNote: 6
Views and downloads (calculated since 01 Feb 2022)
Cumulative views and downloads (calculated since 01 Feb 2022)

Viewed (geographical distribution)

Total article views: 453 (including HTML, PDF, and XML) Thereof 453 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 20 May 2022
Download
Short summary
We present an application of a hybrid deep learning (DL) model on prediction of surface CO in China from 2015 to 2020, which utilizes both convolutional neural networks and long short-term memory neural networks. The performance of DL model is better than a Kalman Filter (KF) system in the training period (2005–2018). Furthermore, DL model demonstrates good temporal extensibility: the mean bias and correlation coefficients are 95.7 ppb and 0.93 in the test period (2019–2020) over E. China.