On This Page:
Welcome
The Earth Prediction Innovation Center (EPIC) and the NOAA Artificial Intelligence for the Numerical Weather Prediction Applications (AI4NWP) Working Group support a community-based, comprehensive Earth modeling system that includes artificial intelligence-based models. The community enables research, development, and contribution opportunities within the broader Weather Enterprise (including government, industry, and academia).
To effectively communicate various AI4NWP activities currently underway across NOAA, we are establishing the NOAA AI-UFS blog. In the NOAA AI-UFS blog posts, we will highlight important developments related to releases of the NOAA training data for AI models, progress on the development of the NOAA-native models, and the NOAA-focused evaluation of the AI4NWP models.
Getting Started
To get started with AI modeling that relies on NOAA initial conditions and datasets, NOAA and CIRA scientists developed this notebook that showcases how to use GraphCast, FourCastNet V2, and PanguWeather with NOAA GFS and GDAS initial conditions https://github.com/
Documentation & User Support
Version
Description
Documentation for the main branch. This includes Jupyter notebooks.
NOAA datasets for training of the AI4NWP models
UFS GEFSv13 replay 0.25 degree subsampled:
noaa-ufs-g…v13replay – Bucket details – Cloud Storage – Google Cloud console
UFS GEFSv13 replay 0.25 degree:
noaa-ufs-g…v13replay – Bucket details – Cloud Storage – Google Cloud console
UFS GEFSv13 replay 1.0 degree:
noaa-ufs-g…v13replay – Bucket details – Cloud Storage – Google Cloud console
NOAA public code base for AI4NWP models
Get Started With Artificial Intelligence
Resources
Public Release Code
NOAA Public Code base for AI4NWP models
For Developers
NOAA datasets for training of the AI4NWP models
Archive of GFS analysis and forecasts (80 members + high res member)
