The Earth Prediction Innovation Center (EPIC), in close collaboration with the Unified Forecast System (UFS) community, has released version 3.0.0 of the UFS Land Data Assimilation (DA) System. This latest release represents a major advancement over the v2.0.0 released in October 2024, by delivering expanded modeling capabilities, broader platform support, and substantial improvements in workflow automation and usability. These enhancements are the result of months of focused development within the land‑DA_workflow and effectively address the evolving needs of the land modeling and DA research community. Land DA v3.0.0 introduces a more flexible and integrated workflow and expands its land-atmosphere modeling capabilities by integrating the ATMosphere-Land (ATML) coupling configuration of the UFS Weather Model. This enables interactive execution of the FV3 UFSATM atmospheric model with the Noah‑MP Land Surface Model (LSM). The v3.0.0 workflow also adds enhanced data‑forcing capabilities through the integration of ECMWF Reanalysis 5th-generation data (ERA5) via the Data ATmosphere Model (DATM) forcing component, along with updates to the ATML physics suite, including FV3_GFS_v17_p8_ugwpv1.
Observation capabilities have been expanded to include Interactive Multi‑sensor Snow and Ice Mapping System (IMS) snow products and an H(x) forward-operator evaluation mode for Soil Moisture Active Passive (SMAP) and Soil Moisture Operational Products System (SMOPS) observations, providing improved tools for forward‑operator evaluation and land DA experimentation. Joint Effort for Data assimilation Integration (JEDI) based H(x) capabilities now allow direct comparison between model background fields and SMOPS observations, enabling SMOPS to be used for evaluating and improving land‑surface initial conditions within this framework. Figure 1 illustrates an example of the JEDI‑based H(x) capability for SMOPS and figure 2 gives an overview of how SMOPS and other land observations integrate into the broader Land‑DA workflow.
Furthermore, Land DA v3.0.0 introduces major improvements in automation, configuration management, and workflow usability. The fully integrated JEDI Configuration Builder (JCB) now auto-generates JEDI input YAML files—with custom overrides supported—producing configurations directly comparable to NOAA’s operational snow DA system. Additional enhancements include a Python‑based experiment setup tool that uses UW tools and Jinja templates to generate experiment directories, configuration files, run‑time metadata, as well as a simplified launch script and improved post‑processing and plotting utilities. The streamlined Jenkins pipelines now run workflow end‑to‑end (WE2E) tests with sample‑analysis checks, improving reproducibility and scientific fidelity. The software stack has been upgraded to spack‑stack v1.9.2, and the system now uses the JEDI Bundle aligned with the Global Data Assimilation System (GDAS) App for consistency with NOAA’s operational DA development.
The workflow runs on all four Tier‑1 Research and Development High‑Performance Computing Systems (RDHPCS), Ursa, Gaea‑C6, Orion, and Hercules, with nightly WE2E tests providing continuous stability verification. Singularity‑based container support has been expanded for HPC and cloud environments. Documentation and community resources have been fully updated. The Land DA v3.0.0 User’s Guide reflects all these enhancements, and a complete suite of test cases is available through the Land‑DA data bucket. Users can engage with developers through the Land DA GitHub Discussions’s Q & A pages. Additionally, the Consortium for Advanced Data Assimilation Research and Education (CADRE) and EPIC have collaboratively developed a complete suite of hands-on training materials centered on the UFS Land DA system, which are publicly available at ucadre.org. Users can also reach out directly through support.epic@noaa.gov.
Looking forward, EPIC and the Land DA team are actively engaged in transitioning the current Land DA system into a full community DA workflow, thereby extending it into a broader community DA infrastructure. Key future developments include Ensemble Land DA, with an initial design leveraging GEOGate, an National Center for Atmospheric Research (NCAR) open-source project intended to serve as the integration layer. Coordination with NOAA’s National Environmental Satellite, Data, and Information Service (NESDIS) will continue to advance full soil-moisture analysis capabilities (Figure 2), including the development of robust water-balance operators. Future platform expansion is also planned to support upcoming NOAA RDHPCS resources.
UFS Land DA v3.0.0 Contributors
This Land DA capability release is a collaboration between the Earth Prediction Innovation Center (EPIC), the Environmental Modeling Center (EMC), the Physical Sciences Laboratory (PSL), the National Environmental Satellite, Data, and Information Service (NESDIS), and the Joint Center for Satellite Data Assimilation (JCSDA). It uses the Joint Effort for Data assimilation Integration (JEDI) software developed by the JCSDA and the Common Community Physics Package (CCPP) developed by the Developmental Testbed Center (DTC). Publicly available data is provided via an AWS S3 bucket established as part of the NOAA Open Data Dissemination (NODD) Program.
Acknowledgements
This release was funded by the NOAA Weather Program Office (WPO)’s Earth Prediction Innovation Center (EPIC) Program, the National Weather Service (NWS) Office of Science and Technology Integration (OSTI) Modeling Programs, and the NOAA Disaster Supplemental Program. Additional support was provided by the NOAA Joint Polar Satellite System (JPSS) Proving Ground and Risk Reduction (PGRR) Program, the NOAA Infrastructure Investment and Jobs Act (IIJA) and the Inflation Reduction Act, the NOAA Climate Program Office (CPO) Modeling, Analysis, Predictions and Projections (MAPP) Program, and the University of Maryland Cooperative Institute for Satellite Earth System Studies (CISESS) seed project.
