19 Aug 2022
19 Aug 2022
Status: this preprint is currently under review for the journal ACP.

Capturing synoptic-scale variations in surface aerosol pollution using deep learning with meteorological data

Jin Feng1, Yanjie Li2, Yulu Qiu1, and Fuxin Zhu1 Jin Feng et al.
  • 1Institute of Urban Meteorology (IUM), China Meteorological Administration (CMA), Beijing, 100089, China
  • 2State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, 100029, China

Abstract. The estimation of daily variations in aerosol concentrations using meteorological data is meaningful and challenging, given the need for accurate air quality forecasts and assessments. In this study, a 3×50-layer spatiotemporal deep learning (DL) model is proposed to link synoptic variations in aerosol concentrations and meteorology, thereby building a “deep” Weather Index for Aerosols (deepWIA). The model was trained and validated using seven years of data and tested in Jan–Apr 2022. The index successfully reproduced the variation in daily PM2.5 observations in China. The coefficient of determination between PM2.5 concentrations calculated from the index and observation was 0.72, with a root-mean-square error of 16.5 µg m−3. DeepWIA performed better than Weather Forecast and Research (WRF)-Chem simulations for eight aerosol-polluted cities in China. The predictive power of the DL model also outperformed reported semi-empirical meteorological indices and machine learning-based PM2.5 concentration retrievals based on aerosol optical depth and visibility observations. The index and the DL model can be used as robust tools for estimating daily variations in aerosol concentrations.

Jin Feng et al.

Status: open (until 27 Oct 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Jin Feng et al.


Total article views: 210 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
165 41 4 210 16 2 1
  • HTML: 165
  • PDF: 41
  • XML: 4
  • Total: 210
  • Supplement: 16
  • BibTeX: 2
  • EndNote: 1
Views and downloads (calculated since 19 Aug 2022)
Cumulative views and downloads (calculated since 19 Aug 2022)

Viewed (geographical distribution)

Total article views: 217 (including HTML, PDF, and XML) Thereof 217 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 28 Sep 2022
Short summary
It is important to use weather data to estimate aerosol concentrations. Here a deep-learning-based weather index for aerosol concentration was developed, linking weather and short-term variations in aerosol concentrations over China. The index provides better performance than chemical transport model simulation and other data-based estimation approaches. It can be used as a robust tool for estimating daily variations in aerosol concentrations.