Learning from machine learning for improved Earth system understanding
Have you ever tried to plan an event a few weeks in advance, and the weather forecast seems to constantly change? You are not alone. Scientists have been working to improve these longer-range — or subseasonal — forecasts for decades. Now, researchers leverage machine learning in this effort, helping us better understand our Earth system and predict it more accurately.
In this Explorer Series conversation, Kirsten Mayer will explore the intersection of machine learning and Earth system prediction. She will discuss how scientists at NSF NCAR are using machine learning to uncover what conditions make forecasts more accurate, with a focus on predictions weeks in advance.
Kirsten Mayer
Kirsten Mayer is a scientist at the NSF National Center for Atmospheric Research (NSF NCAR) in the CGD Laboratory. Her research centers on the identification and exploration of predictability sources across subseasonal to decadal timescales, leveraging machine learning methods. Beyond scientific research, Kirsten has led many AI/ML tutorials, and is currently creating an asynchronous ML training for the Earth System Sciences with fellow scientists and educational specialists across the organization. In addition to her work at NSF NCAR, she is also a member of the World Climate Research Programme Working Group on Subseasonal to Interdecadal Prediction. Kirsten obtained her M.S. and Ph.D. in Atmospheric Science from Colorado State University with Elizabeth Barnes, and a B.S. in Atmospheric and Oceanic Sciences from the University of Wisconsin, Madison.