Researchers from UC San Diego and the Allen Institute for AI have developed a generative AI capable of simulating climate models approximately 25 times faster than traditional methods. This AI can complete a 10-year climate simulation in under three hours, a task that previously took nearly 78 hours on a supercomputer. However, the AI's simulations exhibit about 50% higher biases compared to the original NOAA FV3GFS model, particularly at higher altitudes and polar regions, and it currently only emulates a single scenario without user-defined climate runs. AI
IMPACT This AI advancement could significantly speed up climate change research and prediction, though current biases need to be addressed for broader application.
RANK_REASON The cluster describes a research paper presented at ICML 2024 detailing a new AI model for climate simulation. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — fosstodon.org →
- Allen Institute for Artificial Intelligence
- fv3gfs-wrapper: a Python wrapper of the FV3GFS atmospheric model
- National Oceanic and Atmospheric Administration
- University of California, San Diego
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
