Ai Is Making Whole Earth Simulation Possible why That Matters Mike Pritchard Tedxboston

When responding to extreme weather threats, reliable predictions and warning systems can be the difference between life and death. The world is in need of high-resolution earth system predictions that society can readily interact with. But typical physics simulations of weather and climate are cumbersome — they struggle to scale to desired resolutions and cannot utilize the potential of Earth observations. Artificial intelligence is shattering our notion of what’s possible: planetary simulations can now be done thousands of times faster and more efficiently, unlocking a new science of huge ensemble predictions to study the rarest and most damaging extreme events at high fidelity. Simultaneously, these models are becoming more accessible and user-friendly, democratizing access. The future of AI climate simulation is being born. Mike Pritchard is director of climate simulation research at NVIDIA, and professor and principal investigator of the Computational Clouds and Climate Lab at the University of California, Irvine. Pritchard’s primary research interest is on advancing cloud-resolving weather and climate simulations using physics-informed machine learning. His work blends next-generation global atmospheric simulation algorithms, theoretical climate dynamics, machine learning, and high-performance computing. Pritchard received his Ph.D. at the University of California, San Diego. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at https://www.ted.com/tedx