HomeTechnologyArtificial intelligenceHow Studying the Clouds Can Improve Climate Models

How Studying the Clouds Can Improve Climate Models

How studying the clouds can improve climate models

from Elise Gott
|November 23, 2022

Associate Research Scientist Kara Lamb grew up reading her father’s Scientific American Magazines. She absorbed everything she could about quantum physics and for a time thought she would study it for the rest of her career. But in 2008, after Lamb completed her master’s degree on the subject, the urgency of the climate crisis loomed ever greater. The Intergovernmental Panel on Climate Change has just released its fourth assessment report, and the questions it raised about Earth’s systems were as complex as they were motivating—enough that Lamb chose to redirect her doctorate and subsequent work to the atmospheric sciences. At Columbia Engineering, Lamb is now exploring how machine learning methods can be used to better understand the microphysics of cirrus clouds.

Cirrus clouds over Federal Way, WA. Photo: Ron Clausen via Creative Commons

Cirrus clouds form higher in the atmosphere than almost any other type of cloud — tens of thousands of feet above the ground. They take shape when water vapor is pushed up into the stratosphere by the rise of warm, dry air and then freezes due to the low temperatures. The result is thin, feathery clouds made up of thousands to millions of ice crystals. These ice crystals — and their many different structures, shapes and sizes — influence how cirrus clouds both reflect incoming sunlight and trap outgoing heat from Earth, known as the cloud’s radiation effect.

Lamb is one of the first people to use machine learning to study the shape of the ice crystals in cirrus clouds. With this, she hopes to inform the way scientists account for clouds in climate models.

Must Read
Meta AI Researchers Propose Token Merging (ToMe) to Make Vision Transformers Run Faster

“Even though it seems like a small piece, when you look at the sensitivity of climate models to cloud feedback, there’s a pretty wide spread across the ensemble of climate models out there,” Lamb said.

Machine learning could help to remove this uncertainty by processing and generalizing new knowledge from large amounts of ice crystal data at an unprecedented speed. Traditionally, atmospheric scientists have to come up with a series of mathematical equations, based on physical principles, to describe how a system – in this case a cloud – works. This process can be slow because scientists must iteratively compare the models to observations and then adjust the model’s parameters to improve performance. Machine learning, on the other hand, can be used to learn directly from data; these algorithms build models by “training” on observations.

One of the many ways that machine learning can fit into cloud microphysics research is through pattern recognition. While previous research applying machine learning to ice crystal observations focused on classifying images based on the shape of ice crystals, Lamb plans to take it a step further and actually learn something about the ice growing process directly from the images – with using state-of-the-art methods developed by the scientific machine learning community. Now, three months into the three-year project, Lamb is testing algorithms. After training the model, Lamb estimates she will give it about 12 million different images for this project alone.

Electron microscopy images of individual ice crystals captured by a weather balloon in Billings, OK, and stored in liquid nitrogen for transport back to the lab. Image credit: Nathan Magee (The College of New Jersey) and Jerry Harrington (Penn State University).

“Each data source gives us a different little piece of the picture,” says Lamb. “There we want to use methods from machine learning and statistics to try to understand how to sew all these pieces of information together.”

Must Read
CleanRobotics earns $400,000 EPA fund for AI trash sorting system

Most of the data Lamb uses comes from collaborating with other universities. For example, Kara Sulia at the University of Albany has collected ice crystal observations from several aircraft measurement campaigns into one massive database. At Pennsylvania State University, Jerry Harrington created high-resolution images of ice crystals by using a balloon instrument to collect samples in liquid nitrogen and then freeze them. Other datasets come from large-scale “cloud chamber” experiments where cloud formation is simulated in a lab, allowing scientists to systematically study cirrus cloud formation while testing different aerosol conditions, temperatures and pressures. Lamb uses cloud chamber data from her previous research at the University of Chicago, completed in collaboration with researchers at the Karlsruhe Institute of Technology in Germany.

Kara Lamb, an associate research scientist in Columbia University’s Department of Earth and Environmental Engineering, and a colleague who monitors instruments during cirrus cloud simulation experiments in the Aerosol Interaction and Dynamics in the Atmosphere cloud chamber at Karlsruhe Institute of Technology. Photo: Kara Lam

And yet, despite what all this data shows, most climate models — including those used for the reports of the Intergovernmental Panel on Climate Change — assume that ice crystals are spheres. This is partly because the models are so large-scale that they have to rely on simplified, “computationally efficient” ways to represent complex processes, Lamb explains.

“The model has to be something you can run fast enough to actually get an answer,” she said.

The application of machine learning could somewhat reduce this tension between accuracy and efficiency. As Lamb progresses in using machine learning to analyze ice crystal data, she’s simultaneously collaborating with Marcus van Lier-Walqui at the Columbia Climate School Center for Climate Systems Research and colleagues at the National Center for Atmospheric Research in another project aimed at applying this kind of new knowledge to the Community Earth System Model. The project is in collaboration with the Getting to know the Earth with the Center for Artificial Intelligence and Physics (LEAP). and is the perfect complement to Lamb’s closer investigation of cirrus clouds.

Must Read
Ford takes big charge as autonomy startup Argo AI shuts down, earnings head for low end of guidance

“If we want to understand how important the radiation effects of these clouds are, we need to more accurately represent the shapes of their ice crystals in climate models,” Lamb said. “It’s one of the biggest uncertainties to solve.”

Fortunately, the work of Lamb and her colleagues offers a promising start.



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments