Comparing Machine Learning Models of Raindrop Formation

visha1v1 pts0 comments

Comparing Machine Learning Models of Raindrop Formation - Eos

Close

Search for:

Search

Close<br>About

Special Reports

Topics

Climate

Earth Science

Oceans

Space & Planets

Health & Ecosystems

Culture & Policy

Education & Careers

Opinions

Projects

Postcards From the Field

ENGAGE

Editors’ Highlights

Editors’ Vox

Eos en Español

Eos 简体中文版

Print Archive: 2015–2025

Policy Tracker

Blogs

Research & Developments

The Landslide Blog

Newsletter

Submit to Eos

Skip to content

Eos

Science News by AGU

Support Eos

Sign Up for Newsletter

Menu

Open Search

Search for:

Search

Researchers used simulated “superdroplets” as training data in three machine learning models designed to improve understanding of how droplets in rain clouds coalesce. Here the simulation data are shown superimposed on a public park in Livermore, Calif. Credit: Liam Krauss

Source: Journal of Geophysical Research: Machine Learning and Computation

Raindrops form inside clouds when tiny particles of water collide and stick together, forming larger droplets that eventually fall to Earth. This process is hard to model accurately, with current approaches being either imprecise or computationally intensive. Better simulations of raindrop formation could help improve climate and weather models.

This newsletter rocks.

Get the most fascinating science news stories of the week in your inbox every Friday.

Sign up now

de Jong et al. present and compare three new models of droplet coalescence inside clouds that were created using machine learning techniques: a polynomial-based sparse identification of nonlinear dynamics (SINDy) framework, a neural network–driven time derivative, and a discrete-time autoregressive neural network. Of the three, the SINDy framework (the simplest model) performed best, yielding less uncertainty and better generalization to out-of-sample data than the other two approaches.

The key takeaway, the authors say, is that adding flexibility or complexity to models doesn’t always lead to better performance.

The authors used a machine learning technique called autoencoding to train the models, using data from large eddy simulations that incorporated an approach called the superdroplet method (SDM). By using modeled droplets that represent collections of real droplets, SDM approximates the size distribution and interactions of particles within a cloud more accurately than traditional methods.

The autoencoders successfully reconstructed many aspects of droplet size distributions under coalescence. For example, they were able to effectively reproduce the way mean droplet size increases over time as particles coalesce. They could also accurately simulate bimodal distributions, or datasets with two peaks. In this case, the data’s two peaks indicated that two approximate particle size ranges, often distinguished as being “cloud” or “rain” drops, stand out for being more common than other sizes. However, the autoencoders struggled with re-creating certain features, such as noise or very sharp and narrow peaks in the droplet distributions.

Further work will be necessary to prepare each model for general use, the authors caution, including online testing and the inclusion of other processes, such as condensational growth and evaporation and mixed-phase processes. Future work should focus on pairing simulations with atmospheric observations to impose realistic constraints on models and better tune them for applications in climate and weather modeling, they add. (Journal of Geophysical Research: Machine Learning and Computation, https://doi.org/10.1029/2025JH001103, 2026)

—Nathaniel Scharping (@nathanielscharp), Science Writer

Citation:  Scharping, N. (2026), Comparing machine learning models of raindrop formation, Eos, 107, https://doi.org/10.1029/2026EO260219. Published on 8 July 2026.

Text © 2026. AGU. CC BY-NC-ND 3.0<br>Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

Related

Features from AGU Publications

Research Spotlights

Tracking 20 Years of Productivity in Tidal Wetlands

7 July 20267 July 2026

Editors' Highlights

A Satellite-Based Global Carbon Flux Product is Sensitive to Droughts

8 July 20266 July 2026

Editors' Vox

Small-Scale Indian Ocean Dynamics Underpin Marine Ecology and Climate

4 June 20263 June 2026

About Eos

ENGAGE

Awards

Contact

Advertise

Submit

Career Center

Sitemap

&copy; 2026 American Geophysical Union. All rights reserved

Powered by Newspack

models machine learning search data july

Related Articles