EUVIMEDEUVIMED entdecken
Crossref

Development of a recurrent spatiotemporal deep-learning method coupled with data fusion for correction of hourly ozone forecasts

Jie Li, Ji-Cheng Jang, Yun Zhu, Che-Jen Lin, Shuxiao Wang, Jia Xing, Xinyi Dong, Jinying Li, Bin Zhao, Bingyao Zhang, Yingzhi Yuan

Environmental Pollution · 2023-10 · 335 · 122291

DOI: 10.1016/j.envpol.2023.122291 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Pollution, 2023-10. Verfasst von Jie Li und 10 weiteren Autor:innen. Beteiligte Einrichtung: Guangdong Provincial Key Laboratory of Atmospheric Environment and Pollution Control, School of Environment and Energy, South China University of Tec…. Hauptthemen laut MEDLINE: Ozone, Air Pollutants, Deep Learning, Air Pollution. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 54 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ambient ozone (O3) predictions can be very challenging mainly due to the highly nonlinear photochemistry among its precursors, and meteorological conditions and regional transport can further complicate the O3 formation processes. The emission-based chemical transport models (CTM) are broadly used to predict O3 formation, but they may deviate from observations due to input uncertainties such as emissions and meteorological data, in addition to the treatment of O3 nonlinear chemistry. In this study, an innovative recurrent spatiotemporal deep-learning (RSDL) method with model-monitor coupled convolutional recurrent neural networks (ConvRNN) has been developed to improve O3 predictions of CTM. The RSDL method was first used to build the ConvRNN within a 24-h scale to characterize the spatiotemporal relationships between the monitored O3 data and CTM simulations, and then incorporated the recurrent pattern to achieve 72-h multi-site forecasts based on a pilot study over the Pearl River Delta (PRD) region of China. The results showed that the RSDL method predicted O3 with high accuracy over this case study, with an increase of 27.54% in the correlation coefficient (R) average for all sites as well as an increase in R of 0.14-0.21 for all cities compared to CTM. Moreover, the regional distribution of CTM was further improved by the RSDL predictions with the data fusion technique, which greatly reduced the underpredictions of O3 concentrations, particularly in high O3-level areas (concentrations >160 μg/m3), with a 33.55% reduction in the mean absolute error (MAE).

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OzoneAir PollutantsDeep LearningAir PollutionPilot ProjectsEnvironmental MonitoringChina

Stichwörter der Autor:innen

Air quality modelingData fusionDeep learningOzone photochemistryOzone predictions

Substanzen

OzoneAir Pollutants

Autor:innen und Institutionen

  • Guangdong Provincial Key Laboratory of Atmospheric Environment and Pollution Control, School of Environment and Energy, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou, 510006, Ch…
  • Department of Civil and Environmental Engineering, Lamar University, Beaumont, TX, 77710, USA.
  • State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing, 100084, China.
  • Joint International Research Laboratory of Atmospheric and Earth System Sciences and Institute for Climate and Global Change Research, School of Atmospheric Sciences, Nanjing University, Nanjing, 210023, China.

Zitieren

APA

Li, J., Jang, J. C., Zhu, Y., Lin, C. J., Wang, S., Xing, J., Dong, X., Li, J., Zhao, B., Zhang, B., & Yuan, Y. (2023). Development of a recurrent spatiotemporal deep-learning method coupled with data fusion for correction of hourly ozone forecasts. Environmental Pollution, 335, 122291. https://doi.org/10.1016/j.envpol.2023.122291

Vancouver

Li J, Jang JC, Zhu Y, Lin CJ, Wang S, Xing J, et al. Development of a recurrent spatiotemporal deep-learning method coupled with data fusion for correction of hourly ozone forecasts. Environmental Pollution. 2023;335:122291. doi:10.1016/j.envpol.2023.122291

BibTeX
@article{Li2023,
  author = {Li, Jie and Jang, Ji-Cheng and Zhu, Yun and Lin, Che-Jen and Wang, Shuxiao and Xing, Jia and Dong, Xinyi and Li, Jinying and Zhao, Bin and Zhang, Bingyao and Yuan, Yingzhi},
  title = {Development of a recurrent spatiotemporal deep-learning method coupled with data fusion for correction of hourly ozone forecasts},
  journal = {Environmental Pollution},
  year = {2023},
  volume = {335},
  pages = {122291},
  doi = {10.1016/j.envpol.2023.122291}
}
Kostenlos starten →