Climate models

New project to open years of atmospheric data, improve climate models


A Stanford researcher is leading a project to better understand gravity waves, tiny ripples that contribute to the global circulation of the atmosphere.

Aditi Sheshadri

The first step: estimate the amount of movement carried by gravity waves and make these new estimates available to the research community. And this is no small task.

Since 2013, Loon LLC, a spin-off of Google’s parent company Alphabet, has been launching super pressure balloons into the atmosphere to provide global internet coverage. From their thousands of flights, the balloons have collected high-resolution observations of position, temperature, pressure and wind every 60 seconds – an unprecedented source of information that can be used to better understand gravity waves. .

“It really is, I think, the next frontier in understanding atmospheric dynamics,” said Aditi Sheshadri, atmospheric scientist at Stanford’s School of Earth, Energy & Environmental Sciences (Stanford Earth) and senior researcher at Stanford Earth. DataWave, a new project supported by Schmidt. Future Virtual Earth System Research Institute (VESRI). “There is a lot to be done in terms of observations, high resolution simulations and modeling of gravity waves.”

DataWave aims to make gravity wave statistics from Loon balloon data publicly available in order to assess the sources, propagation and breakdown patterns of gravity waves. From there, researchers will use machine learning to improve both short-term predictions of extreme weather and long-term climate change projections.

“Current representations of gravity waves in climate models involve various simplifying assumptions,” said Sheshadri, assistant professor of Earth System Science. “And yet we know that if we take a climate model and change the gravity wave estimates just a little bit, you fundamentally change the climate. ”

With international collaboration, Sheshadri and his colleagues hope to better understand the impact of gravity waves on large-scale circulation patterns, including the polar vortex, a vortex of freezing air that can bring extreme cold to parts of Europe and the United States for months at a time. Gravity waves are also believed to play a major role in driving the quasi-biennial oscillation, in which a belt of winds blowing above the equator changes direction, impacting the depletion of the ozone layer and surface weather conditions.

DataWave is supported by Schmidt Futures, a philanthropic initiative founded by Eric and Wendy Schmidt that bet early on on great people who make the world a better place. Schmidt Futures’ Virtual Earth System Research Institute (VESRI) aims to dramatically improve the credibility of climate predictions. The funding supports the training of three doctoral students and seven postdoctoral researchers at the intersection of atmospheric dynamics, climate modeling and data science.

Sheshadri is also an assistant professor, out of courtesy, of geophysics, and a member of the center, out of courtesy, at Stanford Woods Institute for the Environment. Project collaborators are affiliated with Goethe-Universität, Northwest Research Associates, UK Met Office, Loon LLC, New York University, Rice University, Laboratoire de Meteorologie Dynamique, Massachusetts Institute of Technology and Max Planck Institute for Meteorology. For more information on the “A Data-Driven Framework for Subnet Scale Gravity Wave Representation to Improve Climate Prediction” project, visit the DataWave website.


Leave a Reply

Your email address will not be published.