![[2602.16821] TopoFlow: Physics-guided Neural Networks for high-resolution air quality prediction](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.16821] TopoFlow: Physics-guided Neural Networks for high-resolution air quality prediction
Summary
TopoFlow introduces a physics-guided neural network for high-resolution air quality prediction, significantly improving accuracy over existing systems.
Why It Matters
This research is crucial as it demonstrates how integrating physical processes into machine learning models can enhance predictive capabilities for air quality, addressing a pressing environmental issue. Improved air quality forecasting can lead to better public health outcomes and inform policy decisions.
Key Takeaways
- TopoFlow achieves a PM2.5 RMSE of 9.71 ug/m3, outperforming existing forecasting systems by 71-80%.
- The model leverages topography and wind direction to enhance pollutant dispersion predictions.
- Performance improvements are consistent across all major pollutants and various forecast lead times.
- The integration of physical knowledge into neural networks can fundamentally advance air quality prediction.
- Forecast errors remain below China's air quality threshold, enabling reliable pollution discrimination.
Computer Science > Machine Learning arXiv:2602.16821 (cs) [Submitted on 18 Feb 2026] Title:TopoFlow: Physics-guided Neural Networks for high-resolution air quality prediction Authors:Ammar Kheder, Helmi Toropainen, Wenqing Peng, Samuel Antão, Jia Chen, Zhi-Song Liu, Michael Boy View a PDF of the paper titled TopoFlow: Physics-guided Neural Networks for high-resolution air quality prediction, by Ammar Kheder and 5 other authors View PDF HTML (experimental) Abstract:We propose TopoFlow (Topography-aware pollutant Flow learning), a physics-guided neural network for efficient, high-resolution air quality prediction. To explicitly embed physical processes into the learning framework, we identify two critical factors governing pollutant dynamics: topography and wind direction. Complex terrain can channel, block, and trap pollutants, while wind acts as a primary driver of their transport and dispersion. Building on these insights, TopoFlow leverages a vision transformer architecture with two novel mechanisms: topography-aware attention, which explicitly models terrain-induced flow patterns, and wind-guided patch reordering, which aligns spatial representations with prevailing wind directions. Trained on six years of high-resolution reanalysis data assimilating observations from over 1,400 surface monitoring stations across China, TopoFlow achieves a PM2.5 RMSE of 9.71 ug/m3, representing a 71-80% improvement over operational forecasting systems and a 13% improvement over state-of...