EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Forecasting Daily Ambient PM2.5 Concentrations in Qingdao City Using Deep Learning and Hybrid Interpretable Models and Analysis of Driving Factors Using SHAP

Zhenfang He, Qingchun Guo iD, Zuhan Zhang, Genyue Feng, Shuaisen Qiao, Zhaosheng Wang iD

Toxics · 30.12.2025 · 14 (1) · 44

DOI: 10.3390/toxics14010044 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Toxics, online erschienen am 30.12.2025. Verfasst von Zhenfang He und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Geography and Environment, Liaocheng University, Liaocheng 252000, China. Die Arbeit wurde bisher 55-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shandong Provincial Natural Science Foundation, State Key Laboratory of Loess Science Foundation, National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the acceleration of urbanization in China, air pollution is becoming increasingly serious, especially PM2.5 pollution, which poses a significant threat to public health. The study employed different deep learning models, including recurrent neural network (RNN), artificial neural network (ANN), convolutional Neural Network (CNN), bidirectional Long Short-Term Memory (BiLSTM), Transformer, and novel hybrid interpretable CNN–BiLSTM–Transformer architectures for forecasting daily PM2.5 concentrations on the integrated dataset. The dataset of meteorological factors and atmospheric pollutants in Qingdao City was used as input features for the model. Among the models tested, the hybrid CNN–BiLSTM–Transformer model achieved the highest prediction accuracy by extracting local features, capturing temporal dependencies in both directions, and enhancing global pattern and key information, with low root Mean Square Error (RMSE) (5.4236 μg/m3), low mean absolute error (MAE) (4.0220 μg/m3), low mean absolute percentage error (MAPE) (22.7791%) and high correlation coefficient (R) (0.9743) values. Shapley additive explanations (SHAP) analysis further revealed that PM10, CO, mean atmospheric temperature, O3, and SO2 are the key influencing factors of PM2.5. This study provides a more comprehensive and multidimensional approach for predicting air pollution, and valuable insights for people’s health and policy makers.

Schlagwörter

Stichwörter der Autor:innen

ANNBiLSTMCNNPM2.5RNNSHAPdeep learningtransformer

Autor:innen und Institutionen

  • School of Geography and Environment, Liaocheng University, Liaocheng 252000, China.
  • Institute of Huanghe Studies, Liaocheng University, Liaocheng 252000, China.
  • State Key Laboratory of Loess Science, Institute of Earth Environment, Chinese Academy of Sciences, Xi'an 710061, China.
  • National Ecosystem Science Data Center, Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China.

ORCID-Profile: Qingchun Guo, Zhaosheng Wang

Förderung

  • Shandong Provincial Natural Science Foundation
  • State Key Laboratory of Loess Science Foundation
  • National Natural Science Foundation of China
  • Liaocheng University Undergraduate Innovation and Entrepreneurship Training Program

Zitieren

APA

He, Z., Guo, Q., Zhang, Z., Feng, G., Qiao, S., & Wang, Z. (2025). Forecasting Daily Ambient PM2.5 Concentrations in Qingdao City Using Deep Learning and Hybrid Interpretable Models and Analysis of Driving Factors Using SHAP. Toxics, 14(1), 44. https://doi.org/10.3390/toxics14010044

Vancouver

He Z, Guo Q, Zhang Z, Feng G, Qiao S, Wang Z. Forecasting Daily Ambient PM2.5 Concentrations in Qingdao City Using Deep Learning and Hybrid Interpretable Models and Analysis of Driving Factors Using SHAP. Toxics. 2025;14(1):44. doi:10.3390/toxics14010044

BibTeX
@article{He2025,
  author = {He, Zhenfang and Guo, Qingchun and Zhang, Zuhan and Feng, Genyue and Qiao, Shuaisen and Wang, Zhaosheng},
  title = {Forecasting Daily Ambient PM2.5 Concentrations in Qingdao City Using Deep Learning and Hybrid Interpretable Models and Analysis of Driving Factors Using SHAP},
  journal = {Toxics},
  year = {2025},
  volume = {14},
  number = {1},
  pages = {44},
  doi = {10.3390/toxics14010044}
}
Kostenlos starten →