Alexander Wikner
Eric and Wendy Schmidt AI in Science Postdoctoral Fellow
RESEARCH:
Alexander Wikner’s research is on estimating the risk of extreme weather by developing and improving AI emulators of the coupled earth system. In the past few years, AI emulators have brought on a “2nd Revolution” in weather forecasting. Emulators trained on the historically observed atmosphere can now produce more skillful forecasts of many atmospheric processes than state-of-the-art, physics-based numerical weather prediction models. At the same time, these emulators are 1,000 to 100,000 times faster to run than these physics-based models. Despite this success, these emulators often struggle to reproduce the dynamics and statistics of very rare, extreme weather events. Alexander aims to leverage techniques from physics, mathematics, and operational weather forecasting to improve the emulator training procedure and the forecasting process to capture the dynamics and risk of extreme weather more accurately.
BIO:
Alexander Wikner completed his Ph.D. in Physics in 2023 at the University of Maryland, where he was advised by Prof. Edward Ott and Prof. Brian R. Hunt. His Ph.D. research focused on developing scalable hybrid machine learning models for forecasting high-dimensional chaotic dynamical systems. Prior to this, Alexander received his B.S. in Physics from Rice University in 2018. Alexander joined UChicago in early 2024 to work with his mentor, Prof. Pedram Hassanzadeh, as a postdoctoral scholar.