Preprints
https://doi.org/10.5194/acp-2021-196
https://doi.org/10.5194/acp-2021-196

  19 Apr 2021

19 Apr 2021

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

Forecasting and Identifying the Meteorological and Hydrological Conditions Favoring the Occurrence of Severe Hazes in Beijing and Shanghai using Deep Learning

Chien Wang Chien Wang
  • Laboratoire d’Aerologie, CNRS and University Paul Sabatier 14 Avenue Edouard Belin, 31400 Toulouse, France

Abstract. Severe haze or low visibility event caused by abundant atmospheric aerosols has become a serious environmental issue in many countries. A framework based on deep convolutional neural networks has been developed to forecast the occurrence of such events in two Asian megacities: Beijing and Shanghai. Trained using time sequential regional maps of meteorological and hydrological variables alongside surface visibility data over the past 41 years, the machine has achieved a good overall accuracy in associating the haze events with favorite meteorological and hydrological conditions. Furthermore, an unsupervised cluster analysis using features with a greatly reduced dimensionality produced by the trained machine has, arguably for the first time, successfully categorized typical regional meteorological-hydrological regimes alongside local quantities associated with haze and non-haze events in the two targeted cities, providing substantial insights to advance our understandings of this environmental extreme.

Chien Wang

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on acp-2021-196', Anonymous Referee #2, 17 May 2021
  • RC2: 'Comment on acp-2021-196', Anonymous Referee #1, 17 May 2021

Chien Wang

Viewed

Total article views: 367 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
277 79 11 367 28 3 3
  • HTML: 277
  • PDF: 79
  • XML: 11
  • Total: 367
  • Supplement: 28
  • BibTeX: 3
  • EndNote: 3
Views and downloads (calculated since 19 Apr 2021)
Cumulative views and downloads (calculated since 19 Apr 2021)

Viewed (geographical distribution)

Total article views: 359 (including HTML, PDF, and XML) Thereof 359 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 29 Jul 2021
Download
Short summary
Haze caused by abundant atmospheric aerosols has become a serious environmental issue in many countries. An innovative deep learning machine has been developed to forecast the occurrence of such events in two Asian megacities: Beijing and Shanghai, and achieved a good overall accuracy. The trained machine has also been used to, arguably for the first time, successfully categorize typical regional meteorological-hydrological regimes associated with haze and non-haze events in the two cities.
Altmetrics