An ensemble machine learning method to retrieve aerosol parameters from ground-based Sun-sky photometer measurements

Li, Qiurui; Sun, Zhongxia; Liu, Meijing; Che, Huizheng; Zheng, Yu; Li, Jing

doi:10.5194/amt-19-2507-2026

Articles | Volume 19, issue 7

https://doi.org/10.5194/amt-19-2507-2026

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

Special issue:

Sun-photometric measurements of aerosols: harmonization, comparisons,...

https://doi.org/10.5194/amt-19-2507-2026

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

Articles | Volume 19, issue 7

Research article

|

16 Apr 2026

Research article |

| 16 Apr 2026

An ensemble machine learning method to retrieve aerosol parameters from ground-based Sun-sky photometer measurements

Qiurui Li, Zhongxia Sun, Meijing Liu, Huizheng Che, Yu Zheng, and Jing Li

Data sets

ERA5 monthly averaged data on pressure levels from 1940 to present Hans Hersbach et al. https://doi.org/10.24381/cds.6860a573

OAA Climate Data Record (CDR) of NASA NOAA LASP Spectral Solar Irradiance (NNLSSI), Version 3 Odele Coddington et al. https://doi.org/10.25921/esjz-1w61

Model code and software

An Ensemble Machine Learning Method to Retrieve Aerosol Parameters from Ground-based Sun-sky Photometer Measurements Qiurui Li https://doi.org/10.5281/zenodo.19398394

Short summary

We present a fast, interpretable machine learning method to retrieve key aerosol parameters from ground-based Sun-sky photometer measurements. Trained on simulated data covering diverse aerosol and atmospheric conditions, ensuring robustness and physical consistency. Applied to real observations, it agrees well with AERONET products and reduces computation time by orders of magnitude, offering a practical tool for monitoring aerosols and their effects on air quality and climate.