Geoscientific Model Development
Geoscientific Model Development
GMD
Articles | Volume 18, issue 11
Articles | Volume 18, issue 11
https://doi.org/10.5194/gmd-18-3509-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-18-3509-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 18, issue 11
Development and technical paper
 | 
13 Jun 2025
Development and technical paper |  | 13 Jun 2025

DustNet (v1): skilful neural network predictions of dust aerosols over the Saharan desert

Trish E. Nowak, Andy T. Augousti, Benno I. Simmons, and Stefan Siegert

Supplement

https://doi.org/10.5194/gmd-18-3509-2025-supplement

Data sets

Pre-processed daily ERA5 and MODIS AOD data (2003 - 2022) ready for use in AI/ML forecasting T. E. Nowak et al. https://doi.org/10.5281/zenodo.10593152

MODIS/Terra Aerosol Cloud Water Vapor Ozone Daily L3 Global 1Deg CMG S. Platnick et al. https://doi.org/10.5067/MODIS/MOD08_D3.006

MODIS/Aqua Aerosol Cloud Water Vapor Ozone Daily L3 Global 1Deg CMG S. Platnick et al. https://doi.org/10.5067/MODIS/MYD08_D3.006

CAMS global atmospheric composition forecasts Copernicus Atmosphere Monitoring Service https://doi.org/10.24381/04a0b097

Model code and software

DustNet - structured data and Python code to reproduce the model, statistical analysis and figures T. E. Nowak et al. https://doi.org/10.5281/zenodo.10722953

Download
Short summary
The DustNet model uses deep neural networks to accurately predict Saharan mineral dust transport in the atmosphere. It offers fast and precise forecasts with predictions achieved in just 2.1 s on a standard computer. This innovative approach outperforms traditional models, which take hours to produce a forecast and use high-energy supercomputers. By making high-quality dust monitoring accessible and efficient, DustNet can improve weather, climate, and air quality forecasts.
Read more
Share