Publication Abstracts
Ko et al. 2025
Ko, J., J.Y. Harrington, K.J. Sulia, V. Przybylo, , and K.D. Lamb, 2025: A machine learning framework for predicting microphysical properties of ice crystals from cloud particle imagery. J. Geophys. Res. Mach. Learn. Comput., 2, no. 4, e2025JH000905, doi:10.1029/2025JH000905.
The microphysical properties of ice crystals are important because they significantly alter the radiative properties and spatiotemporal distributions of clouds, which in turn strongly affect Earth's climate. However, it is challenging to measure key properties of ice crystals, such as mass or morphological features. Here, we present a proof-of-concept framework for predicting three-dimensional (3D) microphysical properties of ice crystals from in situ two-dimensional (2D) imagery. First, we computationally generated synthetic ice crystals using 3D modeling software along with geometric parameters estimated from the 2021 Ice Cryo-Encapsulation Balloon (ICEBall) field campaign. Then, we used synthetic crystals to train machine learning (ML) models to predict effective density ρe, effective surface area Ae, and number of bullets Nb from synthetic rosette imagery. On unseen synthetic images, our ML models accurately predicted ice crystal properties. ResNet-18 performed best, achieving R2 values of 0.99 and 0.98 for ρe and Ae, respectively, and MAE of 0.10 for Nb in single view tasks. Stereo view ResNet-18 further reduced RMSE by 40% for ρe and Ae and reduced MAE by 0.08 for Nb. This work provides a novel ML-driven framework for estimating ice microphysical properties from in situ imagery, which will allow for downstream constraints on microphysical parameterizations, such as the mass-size relationship.
Export citation: [ BibTeX ] [ RIS ]
BibTeX Citation
@article{ko03700v,
author={Ko, J. and Harrington, J. Y. and Sulia, K. J. and Przybylo, V. and van Lier-Walqui, M. and Lamb, K. D.},
title={A machine learning framework for predicting microphysical properties of ice crystals from cloud particle imagery},
year={2025},
journal={Journal of Geophysical Research: Machine Learning and Computation},
volume={2},
number={4},
pages={e2025JH000905},
doi={10.1029/2025JH000905},
}
[ Close ]
RIS Citation
TY - JOUR ID - ko03700v AU - Ko, J. AU - Harrington, J. Y. AU - Sulia, K. J. AU - Przybylo, V. AU - van Lier-Walqui, M. AU - Lamb, K. D. PY - 2025 TI - A machine learning framework for predicting microphysical properties of ice crystals from cloud particle imagery JA - J. Geophys. Res. Mach. Learn. Comput. JO - Journal of Geophysical Research: Machine Learning and Computation VL - 2 IS - 4 SP - e2025JH000905 DO - 10.1029/2025JH000905 ER -
[ Close ]
