Universal ordinary differential equations and the parameterization of warm-rain processes
Abstract
The parameterization of warm-rain formation is a long-standing problem in cloud microphysics because the evolution of bulk cloud variables cannot be derived uniquely from the kinetic collection equation (KCE). Universal ordinary differential equations (UODEs)provide a scientific machine learning framework that combines neural networks with ordinary differential equations and can learn dynamical operators directly from time series. Here, the UODE framework is applied to warm-rain formation by training on trajectories generated with a super-droplet model that approximates the KCE. In contrast to most previous machine-learning approaches, only the prognostic state variables are used for training, without requiring autoconversion or accretion rates as targets. The learned closure accurately reproduces the KCE trajectories within the training domain and, although not explicitly constrained to do so, yields autoconversion and accretion rates that closely resemble those diagnosed from the KCE. Analyzing the learned operator provides new insight into the structure of the widely used Seifert and Beheng (2001) warm-rain parameterization. In particular, it suggests that the empirical suppression of accretion primarily compensates for an overestimation of autoconversion at small rain fractions. Motivated by these findings, a refined analytical formulation is proposed that improves agreement with the KCE while retaining the simplicity of the original parameterization. The results demonstrate that UODEs can serve not only as accurate surrogate models but also as a tool for understanding and improving analytical parameterizations of cloud microphysical processes.
Related
- Parameterization of bottom mixed layer and logarithmic layer heights in central Lake Erie
- Cumulus parameterization controls simulated rainfall more than microphysics, but not at heavy thresholds: a 3 * 2 factorial WRF study over the Western Himalaya
- Towards bridging the gap between Antarctic subglacial boundary conditions and ice-sheet model parameterization
- Probabilistic multi-fidelity climate model parameterization for better generalization
- A performance-based multivariable weighting framework for WRF parameterization-scheme selection
BibTeX
@misc{seifert2026universal,
title = {{Universal ordinary differential equations and the parameterization of warm-rain processes}},
author = {Axel Seifert},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2608.30610},
eprint = {2608.30610},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.30610},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Universal ordinary differential equations and the parameterization of warm-rain processes
AU - Axel Seifert
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2608.30610
UR - https://arxiv.org/abs/2608.30610
AB - The parameterization of warm-rain formation is a long-standing problem in cloud microphysics because the evolution of bulk cloud variables cannot be derived uniquely from the kinetic collection equation (KCE). Universal ordinary differential equations (UODEs)provide a scientific machine learning framework that combines neural networks with ordinary differential equations and can learn dynamical operators directly from time series. Here, the UODE framework is applied to warm-rain formation by training on trajectories generated with a super-droplet model that approximates the KCE. In contrast to most previous machine-learning approaches, only the prognostic state variables are used for training, without requiring autoconversion or accretion rates as targets. The learned closure accurately reproduces the KCE trajectories within the training domain and, although not explicitly constrained to do so, yields autoconversion and accretion rates that closely resemble those diagnosed from the KCE. Analyzing the learned operator provides new insight into the structure of the widely used Seifert and Beheng (2001) warm-rain parameterization. In particular, it suggests that the empirical suppression of accretion primarily compensates for an overestimation of autoconversion at small rain fractions. Motivated by these findings, a refined analytical formulation is proposed that improves agreement with the KCE while retaining the simplicity of the original parameterization. The results demonstrate that UODEs can serve not only as accurate surrogate models but also as a tool for understanding and improving analytical parameterizations of cloud microphysical processes.
ER -