AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment

Vishal Nedungadi, Muhammad Akhtar Munir, Marc Rußwurm, Ron Sarafian, Ioannis N. Athanasiadis, Yinon Rudich, Fahad Shahbaz Khan, Salman Khan
Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation, PMLR 292:46-62, 2025.

Abstract

Air pollution remains a leading global health risk, exacerbated by rapid industrialization and urbanization, contributing significantly to morbidity and mortality rates. In this paper, we introduce AirCast, a novel multi-variable air pollution forecasting model, by combining weather and air quality variables. AirCast employs a multi-task head architecture that simultaneously forecasts atmospheric conditions and pollutant concentrations, improving its understanding of how weather patterns affect air quality. Predicting extreme pollution events is challenging due to their rare occurrence in historic data, resulting in a heavy-tailed distribution of pollution levels. To address this, we propose a novel Frequency-weighted Mean Absolute Error (fMAE) loss, adapted from the class-balanced loss for regression tasks. Informed from domain knowledge, we investigate the selection of key variables known to influence pollution levels. Additionally, we align existing weather and chemical datasets across spatial and temporal dimensions. AirCast’s integrated approach, combining multi-task learning, frequency weighted loss and domain informed variable selection, enables more accurate pollution forecasts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v292-nedungadi25a, title = {AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment}, author = {Nedungadi, Vishal and Munir, Muhammad Akhtar and Ru{\ss}wurm, Marc and Sarafian, Ron and Athanasiadis, Ioannis N. and Rudich, Yinon and Khan, Fahad Shahbaz and Khan, Salman}, booktitle = {Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation}, pages = {46--62}, year = {2025}, editor = {Audebert, Nicolas and Azizpour, Hossein and Barrière, Valentin and Castillo Navarro, Javiera and Czerkawski, Mikolaj and Fang, Heng and Francis, Alistair and Marsocci, Valerio and Nascetti, Andrea and Yadav, Ritu}, volume = {292}, series = {Proceedings of Machine Learning Research}, month = {19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v292/main/assets/nedungadi25a/nedungadi25a.pdf}, url = {https://proceedings.mlr.press/v292/nedungadi25a.html}, abstract = {Air pollution remains a leading global health risk, exacerbated by rapid industrialization and urbanization, contributing significantly to morbidity and mortality rates. In this paper, we introduce AirCast, a novel multi-variable air pollution forecasting model, by combining weather and air quality variables. AirCast employs a multi-task head architecture that simultaneously forecasts atmospheric conditions and pollutant concentrations, improving its understanding of how weather patterns affect air quality. Predicting extreme pollution events is challenging due to their rare occurrence in historic data, resulting in a heavy-tailed distribution of pollution levels. To address this, we propose a novel Frequency-weighted Mean Absolute Error (fMAE) loss, adapted from the class-balanced loss for regression tasks. Informed from domain knowledge, we investigate the selection of key variables known to influence pollution levels. Additionally, we align existing weather and chemical datasets across spatial and temporal dimensions. AirCast’s integrated approach, combining multi-task learning, frequency weighted loss and domain informed variable selection, enables more accurate pollution forecasts.} }
Endnote
%0 Conference Paper %T AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment %A Vishal Nedungadi %A Muhammad Akhtar Munir %A Marc Rußwurm %A Ron Sarafian %A Ioannis N. Athanasiadis %A Yinon Rudich %A Fahad Shahbaz Khan %A Salman Khan %B Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation %C Proceedings of Machine Learning Research %D 2025 %E Nicolas Audebert %E Hossein Azizpour %E Valentin Barrière %E Javiera Castillo Navarro %E Mikolaj Czerkawski %E Heng Fang %E Alistair Francis %E Valerio Marsocci %E Andrea Nascetti %E Ritu Yadav %F pmlr-v292-nedungadi25a %I PMLR %P 46--62 %U https://proceedings.mlr.press/v292/nedungadi25a.html %V 292 %X Air pollution remains a leading global health risk, exacerbated by rapid industrialization and urbanization, contributing significantly to morbidity and mortality rates. In this paper, we introduce AirCast, a novel multi-variable air pollution forecasting model, by combining weather and air quality variables. AirCast employs a multi-task head architecture that simultaneously forecasts atmospheric conditions and pollutant concentrations, improving its understanding of how weather patterns affect air quality. Predicting extreme pollution events is challenging due to their rare occurrence in historic data, resulting in a heavy-tailed distribution of pollution levels. To address this, we propose a novel Frequency-weighted Mean Absolute Error (fMAE) loss, adapted from the class-balanced loss for regression tasks. Informed from domain knowledge, we investigate the selection of key variables known to influence pollution levels. Additionally, we align existing weather and chemical datasets across spatial and temporal dimensions. AirCast’s integrated approach, combining multi-task learning, frequency weighted loss and domain informed variable selection, enables more accurate pollution forecasts.
APA
Nedungadi, V., Munir, M.A., Rußwurm, M., Sarafian, R., Athanasiadis, I.N., Rudich, Y., Khan, F.S. & Khan, S.. (2025). AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment. Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation, in Proceedings of Machine Learning Research 292:46-62 Available from https://proceedings.mlr.press/v292/nedungadi25a.html.

Related Material