Release date: 06/18/2026
NOAA’s Office of Oceanic and Atmospheric Research (OAR)’s Earth Prediction Innovation Center (EPIC), National Weather Service (NWS)’s Office of Modeling and Development (OMD), and the Unified Forecast System (UFS) community are proud to announce the public release of the Community Global Workflow (GW) v1.0.0 (https://github.com/NOAA-EMC/global-workflow). This is the first time EPIC is releasing the Community GW framework, which provides an end‑to‑end workflow designed to run global configurations of medium range numerical weather forecasting for the UFS-Weather Model (WM). It is built to support both research and operational implementations. The Community GW v1.0.0 release can be used on NOAA HPC systems, National Science Foundation (NSF) / National Center for Atmospheric Research (NCAR) Derecho, and native Amazon Web Services (AWS) Elastic Compute Cloud (EC2) instances, enabling broad accessibility for the UFS community.
The GW framework is a unified, configurable workflow system that orchestrates all components required to execute a full global forecast cycle. It is a combination of several model components seamlessly integrated into an end-to-end workflow to pre-process, analyze, generate, post-process, and verify forecast data. GW eliminates the need for manual stitching of these components and ensures consistency between operational forecasts and research experiments. The Community GW v1.0.0 supports Global Forecast System (GFS), Global Ensemble Forecast System (GEFS), Global Chemistry and Aerosol Forecast System (GCAFS), and Seasonal Forecast System (SFS) model configurations. Both the Gridpoint Statistical Interpolation (GSI; atmosphere-only) and Joint Effort for Data Assimilation (DA) Integration (JEDI) systems are included. The Community GW v1.0.0 release builds on the joint operational/development workflow developed at NOAA NWS OMD and extends it for community use through EPIC, enabling new users, researchers, developers, and partners to build and run from among many available configurations.
The GW is an end-to-end framework that orchestrates several modeling components through a unified, configurable workflow which include the UFS-WM for forecast generation, GSI/GSI‑Utils/GSI-monitor for atmospheric DA, GDAS (Global Data Assimilation System) App for transition to the JEDI framework for coupled DA, NOAA Emission and eXchange Unified System (NEXUS) for emissions, GFS‑Utils and UFS‑Utils to run the weather model, and wxflow for workflow management. Forecast verification is performed using OMD’s global verification package, which leverages MET/METplus to compute atmospheric metrics for model runs (as of release, limited to Weather and Climate Operational Supercomputing System (WCOSS) within GW).
Accompanying this release are instructions (AWS Cluster build instructions) on how to create an Amazon Machine Image (AMI)-ID for users to set up their own AWS EC2 instance for running Community GW. These should allow anyone with an AWS account and sufficient funds to run GW, even if they are unable to get access to any of the other supported HPC platforms. The AMI includes all required software pre-installed and automatically mounts the necessary S3 buckets, enabling users to run GW successfully.
A Community GW v1.0.0 User’s Guide, Quick Start Guide, and Contributors Guide are available here to provide further information about this release, including instructions for running and testing the framework. Data files required to run the Community Global Workflow Application are available to the public through OMD’s global data bucket and EPIC’s Community GW v1.0.0 Data Bucket. Interested users may visit EPIC’s Community GW page to access related resources or receive further support through the NOAA-EMC GitHub Discussions Forum by
- Submitting questions through NOAA-EMC GW GitHub Discussions’ Q&A page
- Reporting issues or bugs through GW GitHub Issues page
Contributors:
The inaugural Community Global Workflow v1.0.0 release is a collaboration between the EPIC and NOAA NWS OMD. The code is hosted on GitHub and uses software developed by NOAA NWS OMD, OAR’s EPIC, Global Systems Laboratory (GSL), Physical Sciences Laboratory (PSL), Geophysical Fluid Dynamics Laboratory (GFDL), Air Resources Laboratory (ARL), and partner agencies including Joint Center for Satellite Data Assimilation (JCSDA), Developmental Testbed Center (DTC), National Center for Atmospheric Research (NCAR), and the Sea Ice Model (CICE) Consortium. Publicly available data is provided via AWS S3 buckets established as part of the NOAA Open Data Dissemination (NODD) Program Program.
Future work will expand the capabilities of the Community Global Workflow to include enhanced coupling options, expanded configurations, additional model components, and improved execution pathways.
Acknowledgements:
This release was funded by the NOAA Weather Program Office’s Earth Prediction Innovation Center (EPIC), Subseasonal to Seasonal (S2S), and Joint Technology Transfer Initiative (JTTI) programs, the National Weather Service Office of Science and Technology Integration (OSTI) modeling programs; and the NOAA Disaster Supplemental Program.
