Articles | Volume 15, issue 9
https://doi.org/10.5194/gmd-15-3797-2022
https://doi.org/10.5194/gmd-15-3797-2022
Development and technical paper
 | 
10 May 2022
Development and technical paper |  | 10 May 2022

Development of a deep neural network for predicting 6 h average PM2.5 concentrations up to 2 subsequent days using various training data

Jeong-Beom Lee, Jae-Bum Lee, Youn-Seo Koo, Hee-Yong Kwon, Min-Hyeok Choi, Hyun-Ju Park, and Dae-Gyun Lee

Viewed

Total article views: 2,208 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
1,730 422 56 2,208 45 23 25
  • HTML: 1,730
  • PDF: 422
  • XML: 56
  • Total: 2,208
  • Supplement: 45
  • BibTeX: 23
  • EndNote: 25
Views and downloads (calculated since 15 Nov 2021)
Cumulative views and downloads (calculated since 15 Nov 2021)

Viewed (geographical distribution)

Total article views: 2,208 (including HTML, PDF, and XML) Thereof 2,106 with geography defined and 102 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 28 Mar 2024
Download
Short summary
The predication of PM2.5 has been carried out using a numerical air quality model in South Korea. Despite recent progress of numerical air quality models, accurate prediction of PM2.5 is still challenging. In this study, we developed a data-based model using a deep neural network (DNN) to overcome the limitations of numerical air quality models. The results showed that the DNN model outperformed the CMAQ when it was trained by using observation and forecasting data from the numerical models.