Deep learning representation of the aerosol size distribution

Barahona, Donifan; Breen, Katherine H.; Block, Karoline; Darmenov, Anton

doi:10.5194/gmd-19-2437-2026

Articles | Volume 19, issue 6

https://doi.org/10.5194/gmd-19-2437-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-19-2437-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 19, issue 6

Model description paper

|

26 Mar 2026

Model description paper |

| 26 Mar 2026

Deep learning representation of the aerosol size distribution

Donifan Barahona, Katherine H. Breen, Karoline Block, and Anton Darmenov

Data sets

MERRA-2 inst3_3d_asm_Nv: 3d,3-Hourly,Instantaneous,Model-Level,Assimilation,Assimilated Meteorological Fields V5.12.4 GMAO https://doi.org/10.5067/WWQSXQ8IVFW8

GiOcean Coupled Reanalysis GMAO https://portal.nccs.nasa.gov/datashare/gmao/geos-s2s-3/GiOCEAN_e1/

Cloud condensation nuclei (CCN) numbers derived from CAMS reanalysis EAC4 (Version 1) K. Block https://doi.org/10.26050/WDCC/QUAERERE_CCNCAMS_v1

Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation Lidar Level 2 Aerosol Profile V4-20 CALIPSO https://doi.org/10.5067/CALIOP/CALIPSO/LID_L2_05KMAPRO-STANDARD-V4-20

Model code and software

Initial release of MAMnet D. Barahona and K. H. Breen https://doi.org/10.5281/zenodo.15190121

Short summary

Particulate matter impacts Earth's radiation, clouds, and human health, but modeling their size is challenging due to computational and observational limits. We developed a machine learning model to predict aerosol size distributions, which accurately replicates advanced models and field measurements.