NOAA EAGLE (Experimental AI Global and Limited-Area Ensemble) Global Deterministic and Ensemble Forecasts
InfoSold by: NOAA
Open data
|
Deployed on AWS
<h4>Update</h4>
Effective on December 17, 2025, the <a href="https://www.weather.gov/media/notification/pdf_2025/scn25-89_AIGFS_AIGEFS_and_HGEFS.pdf">NOAA/NWS National Centers for Environmental Prediction (NCEP) implemented three new models</a>: the Artificial Intelligence Global Forecast System (AIGFS), the Artificial Intelligence Global Ensemble Forecast System (AIGEFS), and the Hybrid Global Ensemble Forecast System (HGEFS). The AIGFS/AIGEFS are the operational replacement for EAGLE SOLO/Ensemble. <b>The EAGLE SOLO/Ensemble forecasts hosted here are based on the GraphCastGFS described below</b>. Please note that <b>this</b> bucket for EAGLE SOLO/Ensemble will continue hosting experimental forecasts to support ongoing development of EAGLE <b>global weather, sub-seasonal to seasonal (S2S) forecast models</b> once they are ready.
<br/>
<br/>
The EAGLE SOLO/ensemble forecasts are generated from GraphCast Global Forecast System (GraphCastGFS). GraphCastGFS is an experimental system [...]
Overview
Update
Effective on December 17, 2025, the NOAA/NWS National Centers for Environmental Prediction (NCEP) implemented three new models : the Artificial Intelligence Global Forecast System (AIGFS), the Artificial Intelligence Global Ensemble Forecast System (AIGEFS), and the Hybrid Global Ensemble Forecast System (HGEFS). The AIGFS/AIGEFS are the operational replacement for EAGLE SOLO/Ensemble. The EAGLE SOLO/Ensemble forecasts hosted here are based on the GraphCastGFS described below. Please note that this bucket for EAGLE SOLO/Ensemble will continue hosting experimental forecasts to support ongoing development of EAGLE global weather, sub-seasonal to seasonal (S2S) forecast models once they are ready.The EAGLE SOLO/ensemble forecasts are generated from GraphCast Global Forecast System (GraphCastGFS). GraphCastGFS is an experimental system set up by the National Centers for Environmental Prediction (NCEP) to produce medium range global forecasts. It is built upon Google DeepMind’s pre-trained GraphCast, a Machine Learning Weather Prediction (MLWP) model. The horizontal resolution is a 0.25 degree latitude-longitude grid (about 28 km). The model runs 4 times a day at 00Z, 06Z, 12Z and 18Z cycles. Major atmospheric and surface fields including temperature, wind components, geopotential height, specific humidity, and vertical velocity, are available. The products are 6 hourly forecasts up to 16 days. The data format is GRIB2.
Model upgrade information:
The GraphCastGFS version 1.0 takes two model states as initial conditions (current and 6-hr previous states) from NCEP 0.25 degree GDAS analysis data and runs 37-level GraphCast and 13-level GraphCast-operational with a pre-trained model provided by GraphCast. Unit conversion to the GDAS data is conducted to match the input data required by GraphCast and to generate forecast products consistent with GFS from GraphCastGFS’ native forecast data. The forecast has been uploaded to the bucket since 12Z on February 5, 2024.
The GraphCastGFS version 2.0 made several changes from the GraphCastGFS version 1.0, including: 1) The 37-level model is removed due to the storage restriction and limited accuracy. 2) The 13-level GraphCast ML model was fine-tuned with NCEP’s GDAS analysis data as inputs and ECMWF’s ERA5 data as ground truth. The new weights created from the training are used to create global forecasts. For more details, please refer to NOAA NCEP official note 520 and 522. GraphCastGFS version 2.0 is the base of the EAGLE SOLO/ensemble, whose operational counterparts are AIGFS/AIGEFS v1.0. The EAGLE SOLO version 1.0 forecasts started from 06Z April 24, 2024 and ended at 18Z on December 18, 2025. The EAGLE ensemble version 1.0 forecasts started from 00Z April 29, 2025 and ended at 18Z on December 17, 2025.
Innovations have since been made separately for AIGFS and AIGEFS. To distinguish the two systems, the future development for AIGFS and AIGEFS will be versioned as AIGFSdevX.Y and AIGEFSdevX.Y with experimental forecasts hosted in this bucket.
AIGFSdev2.1 updated loss function in GraphCast from grid-point mean square error to spectral harmonic based mean square error. The loss scaling is also updated for most of the variables. The forecasts started from 00Z Dec 19, 2025 under graphcast.YYYYMMDD/cyc/forecast_13_levels. Additional feature updates will go to EAGLE SOLO test, whose forecast will be under graphcast.YYYYMMDD/cyc/forecast_13_levels_test.
Data information:
The input data generated from the GDAS data is provided under the input/ directory. An example of file names is shown below:
source-gdas_date-2024022000_res-0.25_levels-13_steps-2.nc
The EAGLE SOLO forecast files are under forecasts_13_levels/. There are 128 files (including grib2 index files) under each directory covering 16-day forecasts. An example of a file name is listed below:
graphcastgfs.t00z.pgrb2.0p25.f006
graphcastgfs.t00z.pgrb2.0p25.f006.idx
The EAGLE ensemble forecast files are under EAGLE_ensemble/pmlgefs.YYYYMMDD/cyc/forecasts_13_levels_ICmember_model_X. There are 128 files under each directory covering a16-day forecasts. An example of file name is listed below:
pmlgefsc00.t00z.pgrb2.0p25.f006
pmlgefsc00.t00z.pgrb2.0p25.f006.idx
Please note that this NOAA GraphCastGFS Model was produced using a code package released by Google DeepMind. For information on Google DeepMind, please visit their github page listed in the documentation and license sections of this page.
Features and programs
Open Data Sponsorship Program
This dataset is part of the Open Data Sponsorship Program, an AWS program that covers the cost of storage for publicly available high-value cloud-optimized datasets.
Pricing
This is a publicly available data set. No subscription is required.
Legal
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
AWS Data Exchange (ADX)
AWS Data Exchange is a service that helps AWS easily share and manage data entitlements from other organizations at scale.
Open data resources
Available with or without an AWS account.
- How to use
- To access these resources, reference the Amazon Resource Name (ARN) using the AWS Command Line Interface (CLI). Learn more
- Description
- GraphCast GFS data
- Resource type
- S3 bucket
- Amazon Resource Name (ARN)
- arn:aws:s3:::noaa-nws-graphcastgfs-pds
- AWS region
- us-east-1
- AWS CLI access (No AWS account required)
- aws s3 ls --no-sign-request s3://noaa-nws-graphcastgfs-pds/
- Description
- New data notifications for GraphCast GFS, only Lambda and SQS protocols allowed
- Resource type
- SNS topic
- Amazon Resource Name (ARN)
- arn:aws:sns:us-east-1:709902155096:NewNWSGRAPHCASTGFSObject
- AWS region
- us-east-1
Resources
Vendor resources
Support
Contact
For questions regarding data content or quality, visit the NOAA EMC Github site . For any general questions regarding the NOAA Open Data Dissemination (NODD) Program, email the NODD Team at nodd@noaa.gov .
We also seek to identify case studies on how NOAA data is being used and will be featuring those stories in joint publications and in upcoming events. If you are interested in seeing your story highlighted, please share it with the NODD team by emailing nodd@noaa.gov
Managed By
How to cite
NOAA EAGLE (Experimental AI Global and Limited-Area Ensemble) Global Deterministic and Ensemble Forecasts was accessed on DATE from https://registry.opendata.aws/noaa-nws-graphcastgfs-pds .
License
NOAA's GraphCast GFS products are released under CC0 license. The products in this bucket only contain NOAA produced products and not GraphCast code or training data. For information on the GraphCast code package, please visit the Google DeepMind GraphCast Github License page .
Similar products
S-102 is a data and metadata encoding specification that is part of the S-100 Universal Hydrographic Data Model, an international standard for hydrographic data exchange. This collection of data contains bathymetric surfaces from NOAA/NOS/OCS National Bathymetric Source, for various U.S. coastal and offshore waters and the great lakes. These datasets are encoded as HDF5 files conforming to the S-102 specification.
S-104 is a data and metadata encoding specification that is part of the S-100 Universal Hydrographic Data Model, an international standard for hydrographic data. This collection of data contains water level forecast guidance from NOAA's Global Surge and Tide Operational Forecast System 2-D (STOFS-2D-Global), an operational hydrodynamic nowcast and forecast modeling system for global water level conditions. These datasets are encoded as HDF-5 files conforming to the S-104 specification, and are geospatially subset into individual tiles conforming to the NOAA/OCS Nautical Product Tiling Scheme, with filenames indicating the corresponding NOAA Electronic Navigational Chart (ENC) Cell Identifier. A set of prototype S-104 tiles has been created for the Charleston, SC area for a select model run cycle. Each individual S-104 (HDF-5) file contains all forecast projections from a single mod[...]
This is a 20-year global wave reforecast generated by WAVEWATCH III model (https://github.com/NOAA-EMC/WW3) forced by GEFSv12 winds (https://noaa-gefs-retrospective.s3.amazonaws.com/index.html). The wave ensemble was run with one cycle per day (at 03Z), spatial resolution of 0.25°X0.25° and temporal resolution of 3 hours. There are five ensemble members (control plus four perturbed members) and, once a week (Wednesdays), the ensemble is expanded to eleven members. The forecast range is 16 days and, once a week (Wednesdays), it extends to 35 days. More information about the wave modeling, wave grids and calibration can be found in the WAVEWATCH III regtest ww3ufs1.3 (https://github.com/NOAA-EMC/WW3/tree/develop/regtests/ww3ufs1.3).
The 20 years of reforecast results were analyzed and quality-controlled. Three output types are available
1) Global wave fields, in grib2 format, with several variables including significant wave height, period, direction, and par[...]
S-111 is a data and metadata encoding specification that is part of the S-100 Universal Hydrographic Data Model, an international standard for hydrographic data. This collection of data contains surface water currents forecast guidance from NOAA/NOS Operational Forecast Systems, a set of operational hydrodynamic nowcast and forecast modeling systems, for various U.S. coastal waters and the great lakes. The collection also contains surface current forecast guidance output from the NCEP Global Real-Time Ocean Forecast System (GRTOFS) for some offshore areas. These datasets are encoded as HDF-5 files conforming to the S-111 specification, and are geospatially subset into individual tiles conforming to the NOAA/OCS Nautical Product Tiling Scheme, with filenames indicating the corresponding NOAA Electronic Navigational Chart (ENC) Cell Identifier.
A full set of S-1[...]
NOAA's Climate Data Records (CDRs) are robust, sustainable, and scientifically sound climate records that provide trustworthy information on how, where, and to what extent the land, oceans, atmosphere and ice sheets are changing. These datasets are thoroughly vetted time series measurements with the longevity, consistency, and continuity to assess and measure climate variability and change. NOAA CDRs are vetted using standards established by the National Research Council (NRC).
Climate Data Records are created by merging data from surface, atmosphere, and space-based systems across decades. NOAA’s Climate Data Records provides authoritative and traceable long-term climate records. NOAA developed CDRs by applying modern data analysis methods to historical global satellite data. This process can clarify the underlying climate trends within the data and allows researchers and other users to identify economic and scientific value in these records. NCEI maintains and extends [...]