EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hybrid attention based deep learning for forecasting boundary layer ozone using satellite derived profiles

Shahab S Band, Sultan Noman Qasem, Javad Ramezani, Mohammadreza Abbaspour, Hao-Ting Pai, Brij B Gupta, Zulkefli Mansor, Amir Mosavi

Ecotoxicology and Environmental Safety · 2026-01 · 309 · 119593

DOI: 10.1016/j.ecoenv.2025.119593 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Ecotoxicology and Environmental Safety, 2026-01. Verfasst von Shahab S Band und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information Management, International Graduate School of Artificial Intelligence, National Yunlin University of Science and Technology,…. Hauptthemen laut MEDLINE: Ozone, Deep Learning, Air Pollutants, Environmental Monitoring. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ground level ozone is considered a major air pollutant. It is formed when nitrogen oxides and volatile organic compounds react under sunlight. It harms human health and damages plants and materials. It also contributes to climate change. It is a photochemically formed compound that is extremely hazardous to the environment and human health. Proper forecasting of the boundary layer ozone has been a challenge due to the nonlinearly related to both meteorological and chemical conditions and the scarcity of fine-scale vertical ozone patterns. This study uses the OMPROFOZ ozone profile product which is a product of the Ozone Monitoring Instrument (OMI) on the Aura satellite to estimate the ozone concentrations in the boundary layer. A set of deep learning models, i.e., RNN, CNN, GRU, LSTM, and hybrid forms i.e., GRU-CNN and LSTM-CNN, is evaluated to benchmark forecasting accuracy. The first, ConvBiGRU-AttentionNet, integrates attention mechanisms within a convolutional gated recurrent structure. The second, EMD-ConvBiGRU-AttentionNet, adds Empirical Mode Decomposition to extract multi-scale temporal features before modeling. The proposed models outperform conventional methods across metrics such as RMSE, MAE, R2, and skill scores. EMD-ConvBiGRU-AttentionNet achieves the highest prediction accuracy. Visual analyses, i.e., residual plots, cumulative error distributions, and attention maps, confirm the capacity of the model to capture spatio-temporal patterns in atmospheric data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OzoneDeep LearningAir PollutantsEnvironmental MonitoringForecasting

Stichwörter der Autor:innen

Artificial intelligenceAttention mechanismBig dataBoundary layer ozoneData ScienceDeep learningEmpirical Mode DecompositionMachine learningRemote sensing

Substanzen

OzoneAir Pollutants

Autor:innen und Institutionen

  • Department of Information Management, International Graduate School of Artificial Intelligence, National Yunlin University of Science and Technology, Douliu, Taiwan. Electronic address: shahab@yuntech.edu.tw.
  • Computer Science Department, College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia.
  • Department of Information Management, National Yunlin University of Science and Technology, Yunlin, Taiwan.
  • Department of Big Data Business Analytics, National Pingtung University, Pingtung, Taiwan.
  • Department of Computer Science and Information Engineering, Asia University, Taichung 413, Taiwan; Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan,; Symbiosi…
  • Research Center for Software Technology and Management, Faculty of Information Science, and Technology, Universiti Kebangsaan Malaysia, Bandar Baru Bangi, Selangor Darul Ehsan 43600, Malaysia.

Förderung

  • National Science and Technology Council

Zitieren

APA

Band, S. S., Qasem, S. N., Ramezani, J., Abbaspour, M., Pai, H. T., Gupta, B. B., Mansor, Z., & Mosavi, A. (2026). Hybrid attention based deep learning for forecasting boundary layer ozone using satellite derived profiles. Ecotoxicology and Environmental Safety, 309, 119593. https://doi.org/10.1016/j.ecoenv.2025.119593

Vancouver

Band SS, Qasem SN, Ramezani J, Abbaspour M, Pai HT, Gupta BB, et al. Hybrid attention based deep learning for forecasting boundary layer ozone using satellite derived profiles. Ecotoxicology and Environmental Safety. 2026;309:119593. doi:10.1016/j.ecoenv.2025.119593

BibTeX
@article{Band2026,
  author = {Band, Shahab S and Qasem, Sultan Noman and Ramezani, Javad and Abbaspour, Mohammadreza and Pai, Hao-Ting and Gupta, Brij B and Mansor, Zulkefli and Mosavi, Amir},
  title = {Hybrid attention based deep learning for forecasting boundary layer ozone using satellite derived profiles},
  journal = {Ecotoxicology and Environmental Safety},
  year = {2026},
  volume = {309},
  pages = {119593},
  doi = {10.1016/j.ecoenv.2025.119593}
}
Kostenlos starten →