AirMPA improves global air pollution forecasts for five days
AirMPA: A Meteorology-to-Pollution Adapter for Global Air Quality Forecasting
Artificial Intelligence
Summary
Predicting air pollution accurately is important for keeping people safe from harmful exposure. The authors created AirMPA, a model that separately processes weather and pollution data to better forecast pollution levels worldwide up to five days ahead. This approach lets the model focus on complex pollution behaviors and how pollutants move globally. Testing showed AirMPA forecasts more accurately than existing systems across most pollutants and time frames.
What this means in practice
- •For environmental monitoring teams: Produce more accurate multi-day forecasts of various air pollutants using separate weather inputs to support timely public health advisories.
- •For urban planners: Integrate improved pollution forecasts into city planning tools to better manage emission sources and protect population health.
Authors
Yiheng Wang, Kai Zheng, Yuetan Lin, Fanglu Fan, Chenliang Tao, Guochao Chen, Hongliang Zhang, Hao Li
Abstract
Accurate real-time forecasting of atmospheric pollutants is essential for reducing exposure risks and improving public health. Traditional numerical forecasting systems suffer from incomplete parameterizations, uncertain inputs, and high computational cost. Existing deep learning approaches improve efficiency, but often entangle meteorological and pollutant variables within unified forecasting frameworks, limiting flexibility and pollutant-specific representation learning. To address these challenges, we propose AirMPA, a decoupled autoregressive framework for meteorology-to-pollution forecasting. AirMPA predicts 13 atmospheric pollutants at 0.4° spatial resolution and 12-h temporal resolution using meteorological fields, emission priors, and historical pollutant concentrations. The model is built on an improved Swin-Transformer architecture with zonal release attention and dual-branch fusion, enabling more effective modeling of global transport continuity and heterogeneous pollutant dynamics. During training, AirMPA uses the fifth-generation atmospheric reanalysis from the European Centre for Medium-Range Weather Forecasts (ERA5) as meteorological forcing, while during inference it directly ingests forecast fields from external meteorological models, enabling stable autoregressive prediction up to 5 days ahead. Experiments show that AirMPA consistently outperforms Aurora and surpasses the Copernicus Atmosphere Monitoring Service (CAMS) in approximately 90% of pollutant-lead-time evaluation cases over the 5-day forecast horizon, while outperforming CAMS on all variables beyond 48 h. These results demonstrate the potential of decoupled meteorology-to-pollution forecasting for global air quality prediction.