Interpretable Air Pollution Forecasting by Physics-Guided Spatiotemporal Decoupling

TL;DR

This paper introduces a physics-guided, interpretable spatiotemporal model for air pollution forecasting that decomposes pollutant behavior into transparent modules, outperforming existing methods while maintaining interpretability.

Contribution

The study presents a novel physics-guided, interpretable framework that combines a transport kernel and attention mechanism for improved air pollution prediction.

Findings

Outperforms state-of-the-art baselines across multiple horizons

Provides transparent, physics-informed decomposition of pollutant dynamics

Demonstrates reliable real-world air-quality management potential

Abstract

Accurate and interpretable air pollution forecasting is crucial for public health, but most models face a trade-off between performance and interpretability. This study proposes a physics-guided, interpretable-by-design spatiotemporal learning framework. The model decomposes the spatiotemporal behavior of air pollutant concentrations into two transparent, additive modules. The first is a physics-guided transport kernel with directed weights conditioned on wind and geography (advection). The second is an explainable attention mechanism that learns local responses and attributes future concentrations to specific historical lags and exogenous drivers. Evaluated on a comprehensive dataset from the Stockholm region, our model consistently outperforms state-of-the-art baselines across multiple forecasting horizons. Our model's integration of high predictive performance and spatiotemporal interpretability provides a more reliable foundation for operational air-quality management in real-world applications.