EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

72-hour real-time forecasting of ambient PM2.5 by hybrid graph deep neural network with aggregated neighborhood spatiotemporal information

Mengfan Teng, Siwei Li, Jia Xing, Chunying Fan, Jie Yang, Shuo Wang, Ge Song, Yu Ding, Jiaxin Dong, Shansi Wang

Environment International · 2023-06 · 176 · 107971

DOI: 10.1016/j.envint.2023.107971 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environment International, 2023-06. Verfasst von Mengfan Teng und 9 weiteren Autor:innen. Beteiligte Einrichtung: Hubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, W…. Hauptthemen laut MEDLINE: Air Pollutants, Air Pollution. Die Arbeit wurde bisher 60-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The observation-based air pollution forecasting method has high computational efficiency over traditional numerical models, but a poor ability in long-term (after 6 h) forecasting due to a lack of detailed representation of atmospheric processes associated with the pollution transport. To address such limitation, here we propose a novel real-time air pollution forecasting model that applies a hybrid graph deep neural network (GNN_LSTM) to dynamically capture the spatiotemporal correlations among neighborhood monitoring sites to better represent the physical mechanism of pollutant transport across the space with the graph structure which is established with features (angle, wind speed, and wind direction) of neighborhood sites to quantify their interactions. Such design substantially improves the model performance in 72-hour PM2.5 forecasting over the whole Beijing-Tianjin-Hebei region (overall R2 increases from 0.6 to 0.79), particularly for polluted episodes (PM2.5 concentration > 55 µg/m3) with pronounced regional transport to be captured by GNN_LSTM model. The inclusion of the AOD feature further enhances the model performance in predicting PM2.5 over the sites where the AOD can inform additional aloft PM2.5 pollution features related to regional transport. The importance of neighborhood site (particularly for those in the upwind flow pathway of the target area) features for long-term PM2.5 forecast is demonstrated by the increased performance in predicting PM2.5 in the target city (Beijing) with the inclusion of additional 128 neighborhood sites. Moreover, the newly developed GNN_LSTM model also implies the "source"-receptor relationship, as impacts from distanced sites associated with regional transport grow along with the forecasting time (from 0% to 38% in 72 h) following the wind flow. Such results suggest the great potential of GNN_LSTM in long-term air quality forecasting and air pollution prevention.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Air PollutantsAir PollutionParticulate MatterEnvironmental MonitoringForecastingNeural Networks, Computer

Stichwörter der Autor:innen

AODHybrid graph deep neural networkPM(2.5)Physical mechanism

Substanzen

Air PollutantsParticulate Matter

Autor:innen und Institutionen

  • Hubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China.
  • Hubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China; Hubei Luojia Laboratory, Wuhan University, Wuhan 4…
  • Department of Civil and Environmental Engineering, the University of Tennessee, Knoxville, TN 37996, USA.
  • School of Systems Science, Beijing Normal University, Beijing 100875, China.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Teng, M., Li, S., Xing, J., Fan, C., Yang, J., Wang, S., Song, G., Ding, Y., Dong, J., & Wang, S. (2023). 72-hour real-time forecasting of ambient PM2.5 by hybrid graph deep neural network with aggregated neighborhood spatiotemporal information. Environment International, 176, 107971. https://doi.org/10.1016/j.envint.2023.107971

Vancouver

Teng M, Li S, Xing J, Fan C, Yang J, Wang S, et al. 72-hour real-time forecasting of ambient PM2.5 by hybrid graph deep neural network with aggregated neighborhood spatiotemporal information. Environment International. 2023;176:107971. doi:10.1016/j.envint.2023.107971

BibTeX
@article{Teng2023,
  author = {Teng, Mengfan and Li, Siwei and Xing, Jia and Fan, Chunying and Yang, Jie and Wang, Shuo and Song, Ge and Ding, Yu and Dong, Jiaxin and Wang, Shansi},
  title = {72-hour real-time forecasting of ambient PM2.5 by hybrid graph deep neural network with aggregated neighborhood spatiotemporal information},
  journal = {Environment International},
  year = {2023},
  volume = {176},
  pages = {107971},
  doi = {10.1016/j.envint.2023.107971}
}
Kostenlos starten →