AUTHOR=Xie Zi-Ang , Chow Chee-Onn , Chuah Joon Huang , Raymond Wong Jee Keen 

TITLE=Multi-pollutant air quality forecasting using bidirectional attention and multi-scale temporal networks

JOURNAL=Frontiers in Environmental Science

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2025.1623630

DOI=10.3389/fenvs.2025.1623630

ISSN=2296-665X

ABSTRACT=IntroductionAccurate multi-pollutant forecasting is vital for urban governance and public health. Existing deep models struggle to capture multi-scale temporal dynamics and synergistic cross-pollutant relations.MethodsWe propose an Enhanced Bidirectional Attention Multi-scale Temporal Network (EBAMTN) that combines a multi-scale TCN with linear attention, a two-layer BiLSTM augmented by multi-head self-attention, and a gated fusion layer. Under a multi-task paradigm, the backbone jointly learns shared temporal representations and outputs PM2.5 and PM10 via task-specific heads.ResultsUsing hourly data from Guangzhou, Beijing, and Chengdu, EBAMTN achieved R2 > 0.94 for both pollutants while maintaining low errors (e.g., PM2.5 MAE≈2.03, RMSE≈2.94; PM10 MAE≈3.44, RMSE≈4.99). Confidence-interval analyses and scatter plots indicate strong trend tracking and robustness, with remaining challenges mainly at sharp peaks.DiscussionThe integration of multi-scale convolutions, bidirectional memory, attention, and gated fusion improves accuracy, interpretability, and generalization. The lightweight design (≈2.1M parameters; ∼ 13.2 ms/sample) supports real-time and edge deployment. Overall, EBAMTN offers a scalable, interpretable solution for multi-pollutant forecasting in complex urban settings.