AIFS-COMPO: A Global Data-Driven Atmospheric Composition Forecasting System
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917383662206976 |
|---|---|
| author | Harder, Paula Flemming, Johannes Alexe, Mihai Mertes, Gert Raoult, Baudouin Chantry, Matthew |
| author_facet | Harder, Paula Flemming, Johannes Alexe, Mihai Mertes, Gert Raoult, Baudouin Chantry, Matthew |
| contents | We introduce AIFS-COMPO, a skilful medium-range data-driven global forecasting system for aerosols and reactive gases. Building on the ECMWF Artificial Intelligence Forecast System (AIFS), AIFS-COMPO employs a transformer-based encoder-processor-decoder architecture to jointly model meteorological and atmospheric composition variables. The model is trained on Copernicus Atmosphere Monitoring Service (CAMS) reanalysis, analysis, and forecast data to learn the coupled dynamics of weather, emissions, transport, and atmospheric chemistry. We evaluate AIFS-COMPO against a range of atmospheric composition observations and compare its performance with the operational CAMS global forecasting system IFS-COMPO. The results show that AIFS-COMPO achieves comparable or improved forecast skill for several key species while requiring only a fraction of the computational resources. Furthermore, the efficiency of the approach enables forecasts beyond the current operational horizon, demonstrating the potential of AI-based systems for fast and accurate global atmospheric composition prediction. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_03300 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AIFS-COMPO: A Global Data-Driven Atmospheric Composition Forecasting System Harder, Paula Flemming, Johannes Alexe, Mihai Mertes, Gert Raoult, Baudouin Chantry, Matthew Atmospheric and Oceanic Physics Artificial Intelligence We introduce AIFS-COMPO, a skilful medium-range data-driven global forecasting system for aerosols and reactive gases. Building on the ECMWF Artificial Intelligence Forecast System (AIFS), AIFS-COMPO employs a transformer-based encoder-processor-decoder architecture to jointly model meteorological and atmospheric composition variables. The model is trained on Copernicus Atmosphere Monitoring Service (CAMS) reanalysis, analysis, and forecast data to learn the coupled dynamics of weather, emissions, transport, and atmospheric chemistry. We evaluate AIFS-COMPO against a range of atmospheric composition observations and compare its performance with the operational CAMS global forecasting system IFS-COMPO. The results show that AIFS-COMPO achieves comparable or improved forecast skill for several key species while requiring only a fraction of the computational resources. Furthermore, the efficiency of the approach enables forecasts beyond the current operational horizon, demonstrating the potential of AI-based systems for fast and accurate global atmospheric composition prediction. |
| title | AIFS-COMPO: A Global Data-Driven Atmospheric Composition Forecasting System |
| topic | Atmospheric and Oceanic Physics Artificial Intelligence |
| url | https://arxiv.org/abs/2604.03300 |