GRADUATE STUDENT OPPORTUNITIES FOR UFS FEEDBACK
Would you be willing to provide feedback on your experience with the UFS? UFS university collaborators introduced the idea of the “Graduate Student Test” as a measure of how successful the UFS project is in opening its code and development processes to the broader community. It is based on the belief that a graduate student should be able to easily get, change and run UFS code, test it for correct operation, and use standard diagnostic packages to evaluate results. Starting in 2019, the UFS project began defining specific Graduate Student Tests and collecting feedback through questionnaires. If you are a graduate student or postdoc please consider participating, and if you are a university researcher, consider encouraging your students to participate. Sign up for a graduate student test here.
FIRST UFS RELEASE: THE UFS MEDIUM-RANGE WEATHER APPLICATION 1.0
The first UFS application was released on March 11, 2020! The UFS Medium-Range Weather Application 1.0 targets predictions of global atmospheric behavior out to about two weeks. It includes a prognostic atmospheric model, pre- and post-processing, and a community workflow that greatly simplifies building and running the code. The release is distributed through GitHub and is designed to be a prediction system that the research community can run and improve. The release team was led by NOAA and NCAR partners and involved dozens of contributors from a diverse set of organizations. Through their efforts, the Medium-Range Weather Application has a portable code base, documentation, and a user forum. Read more about the release here.
UFS AT THE AMERICAN METEOROLOGICAL SOCIETY ANNUAL MEETING
The UFS will be featured in presentations and events throughout the 100th Annual American Meteorological Society meeting, to be held in Boston, Massachusetts from January 12-20, 2020. A key event is the 30th Conference on Weather Analysis and Forecasting (WAF)/26th Conference on Numerical Weather Prediction (NWP). This multi-day conference includes talks on planning and testing for data assimilation in the next major release of the Global Forecast System, convection allowing ensemble prediction, the effects of varying vertical resolution and microphysics in the regional atmosphere, and other aspects of the evolving suite of UFS applications. Other events that highlight UFS challenges and achievements are the 10th Conference on Transition of Research to Operations and the Eighth Symposium on Building a Weather-Ready Nation. There are many other events in which UFS is featured! See the full listing of UFS-related events at the AMS Annual Meeting.
COMING SOON: THE UFS MEDIUM-RANGE WEATHER APPLICATION 1.0 RELEASE!
The first release of a UFS application is coming in early 2020! The release will be of the UFS Medium-Range Weather Application 1.0, which targets predictions of global atmospheric behavior out to about two weeks. It includes a prognostic atmospheric model, pre- and post-processing, and a community workflow. The release will be distributed through GitHub and is designed to be code that the research community can run and improve. The release will not yet include data assimilation or a verification package as part of the workflow, and will not be coupled to ocean, wave or ice models, although these capabilities are planned for future releases of this application. Other UFS applications (Subseasonal to Seasonal, Short-Range Weather, more) are also on the way. A description of the release and timeline for the release have more information, and will be updated as the release gets closer.
HIERARCHICAL SYSTEM DEVELOPMENT FOR THE UFS
Hierarchical system development (HSD) is the ability to engage in development and testing at multiple levels of complex prediction software, essential capabilities for advancing UFS. It is critical for supporting research because it enables researchers to have multiple entry points into development that reflect their interests. Those interests may include aspects of specific system components such as atmospheric physics, ocean and ice dynamics, or data assimilation for land models and other individual Earth system components. It may also include more integrative aspects such as coupled modeling or data assimilation. HSD is likewise critical for operations because both localized and integrative, coupled processes must be improved to develop excellence in forecasts. UFS collaborators developed a short article that outlines the types of tests and development practices that will enable HSD to become an integral part of UFS.
UFS SHORT-RANGE WEATHER WORKSHOP
The UFS Short-Range Weather / Convection Allowing Model (CAM) Application Team (AT) and related stakeholders gathered for a workshop on 9-10 September 2019 at the NOAA David Skaggs Research Center in Boulder, CO including representatives from EMC, GSD, NSSL, AOML, DTC, JCSDA and OU/CAPS. The two-day workshop focused on the development status of the Stand Alone Regional (SAR) FV3 model application including tests with data assimilation and various physics suites. The CAM AT is focused on helping simplify the NCEP operational regional model production suite through unification around the SAR FV3 and Joint Effort for Data assimilation Integration (JEDI) systems with an eye towards building Warn On Forecast and Rapid Refresh Forecast Systems (WoFS and RRFS) UFS applications to meet short-range weather / convection allowing model needs across the weather enterprise. The CAM AT is also working to provide a research-ready SAR model capability in the form of public releases to facilitate community engagement in future research, development and transitions to operations.
MAJOR GFS UPGRADE
An upgrade to the Global Forecast System (GFSv15), a key part of the UFS, is now in operations. The Model Evaluation Group (MEG), part of NOAA’s Environmental Modeling Center, conducted a rigorous evaluation. There are two main changes. The first is the FV3 dynamical core – the part of the software that represents the equations of fluid motion – which was selected for its scientific integrity and computational performance. The second is a new microphysics scheme, which is expected to improve the modeling of clouds and precipitation. The evaluation involves comprehensive statistical testing and many retrospective tests, which look at the model’s ability to simulate previous weather events. During extensive side-by-side testing the upgraded GFS outperformed the older version in predicting cyclone track and intensity forecasts.
NOAA-NCAR PARTNERSHIP ON RESEARCH AND OPERATIONAL INFRASTRUCTURE
UFS requires a substantial amount of infrastructure software as well as scientific code. This infrastructure performs functions such as processing data inputs and outputs, enabling software components to be exchanged and coupled together, and ensuring that a forecast workflow can be reproduced on different computers. In January 2019, NOAA and NCAR signed a Memorandum of Agreement to create a partnership on infrastructure development. This partnership will leverage and extend community software developed by these research and operational centers together with partner organizations such as NASA, the Department of Defense, and universities. Sharing infrastructure software across the community means that it is easier for researchers to run operational software and for new science to be transferred to operations.
PLANNING FOR THE FUTURE
UFS will address forecast applications ranging from immediate threats, like thunderstorms and tornadoes, to long-term climate patterns like El Nino. GFS with the FV3/microphysics upgrade has the ability to improve the simulation of small-scale and local processes, like the physics of clouds, the convective movements of air, and the boundary between the atmosphere and land. There is active development on a regional-scale version of GFS, which includes advances in these cloud and boundary layer schemes. Already, these storm-scale experiments show a path forward to more accurate forecasts of dangerous weather events. UFS also supports applications that couple the GFS with models, for example, of the ocean, ice, and waves. Coupled models are essential for wave-storm surge interactions, sub-seasonal to seasonal predictions, and other kinds of forecasts, and have been shown to have important impacts on global weather forecasts for time spans of less than 10 days. Many coupled configurations are already being tested in UFS.