Presenter(s):Jonathan Poterjoy, University of Maryland
Designing fully non-parametric state and parameter estimation methods for the UFS

Environmental prediction centers routinely use numerical models to forecast various components of the Earth system, ranging from the geosphere to the atmosphere. Despite the diversity in applications, these models share two common characteristics, namely, they rely on physical laws to govern the time-rate-of-change of prognostic state variables and data assimilation guides how measurements inform estimates of initial conditions, boundary conditions, or unknown model parameters. The relative skill of model forecasts (e.g., comparisons of two or more global weather prediction systems) is often dominated by algorithmic choices made for data assimilation, which can be further traced back to assumptions made in priors and likelihoods used when formulating such methods from Bayes’ theorem. This seminar will focus more narrowly on atmospheric applications and discuss major obstacles to performing data assimilation with current high-resolution weather models. We will discuss the limitations of Gaussian approximations that are typically made for prior probability densities and exploit the flexibility of “particle filters,” which avoid such approximations. We will also explore the use of kernel-estimated likelihood functions trained from data accumulated during data assimilation and regime-dependent likelihoods represented using kernel embeddings of conditional distributions. These approaches allow for non-Gaussian estimates of likelihood functions that can be used directly by particle filters—or used to compute expectations of bias and error covariance for Gaussian-based data assimilation. Of equal importance, this methodology scales well for high dimensional applications, which are needed for capturing error dependence across observations. The end result is a data assimilation approach that is entirely “non-parametric” in that none of the required error distributions follow specific “shapes” determined by parameters (e.g., mean and covariance for a Gaussian). Beyond weather applications, we will discuss the theoretical advantages of fully non-parametric data assimilation for coupled Earth system models, which require accurate state estimates for the ocean, sea ice, and other component systems.

Dr. Jon Poterjoy is an Assistant Professor at the University of Maryland (UMD) Department of Atmospheric and Oceanic Science (AOSC), where he leads a research group that focuses on uncertainty quantification for geophysical systems, new data assimilation methodology, and coupled atmosphere/ocean/cryosphere modeling. Before joining UMD, Jon was a National Research Council postdoc fellow at the NOAA Atlantic Oceanic and Meteorological Laboratory and an Advanced Study Program postdoc fellow at the National Center for Atmospheric Research. Jon also held a postdoc position at the University of Oklahoma/NOAA Cooperative Institute for Mesoscale Meteorological Studies, where he worked with scientists at the NOAA National Severe Storms Laboratory. He has a Ph.D. in Meteorology from the Pennsylvania State University and a B.S. in Applied Mathematics and Meteorology from Millersville University.

Watch on Youtube
Presentation Slides

Presenter(s): Jun Wang, NWS/NCEP/EMC
Model infrastructure development in UFS weather model

The Unified Forecast System weather model is the community-based coupled earth science modeling system. Through the collaboration of NOAA laboratories and the broad research community, the coupling infrastructure has been developed with new model components integrated into the system and the coupling capability set up to support various configurations. Currently the UFS coupled model consists of FV3 dynamical core with the Common Community Physics Package (CCPP) for the atmosphere, MOM6 and HYCOM for the ocean, CICE6 for sea ice, WW3 for ocean waves, NoahMP for land, GOCART and CMAQ for aerosol and chemistry and the Community Mediator for Earth Prediction Systems (CMEPS) based on ESMF NUOPC for coupling framework. The UFS weather model supports the Hurricane Analysis and Forecast System (HAFS) v1, Regional Air Quality Model (AQM)v7, and upcoming Rapid Refresh Forecast System (RRFS) v1, Global Forecast System (GFSv17) and Global Ensemble Forecast System (GEFSv13) implementations.

In this presentation, an overview of general model infrastructure for the UFS coupled model will be provided. Major infrastructure achievements will be presented. New approaches that were implemented to improve the computational performance will also be discussed.

Jun Wang has been at the Environmental Modeling Center since 2002 working on numerical modeling and system architecture development. She developed the atmosphere ocean coupler for NCEP’s Climate Forecast Model version 2 and later worked on various earth modeling systems including Global Forecast Model, whole atmosphere space weather model, and global aerosol models and UFS. Currently she is the team lead and member developing the UFS weather model infrastructure to support operational implementations and the research community.

Watch on Youtube
Presentation Slides

Presenter(s): Barry Baker, NOAA Air Resources Laboratory
The history and development of the NOAA FENGSHA dust emission scheme.

Aerosols have both direct and indirect effects on meteorology, atmospheric chemistry, human health, and ultimately the global energy budget with dust being a major contributor to the atmospheric aerosol burden. A great effort has been made to characterize the sources and mobilization of dust, however, current models still show large uncertainty as the modeling of mineral dust in the atmosphere is complex. The FENGSHA dust emission model, implemented into the operational NOAA National Air Quality Forecast Capability (NAQFC), and the new Global Ensemble Forecast System with Aerosols (GEFS-Aerosols), is a flexible emission model capable of predicting dust emissions across forecast scales. Here we will discuss the history and development of the FENGSHA dust emission scheme and its applications in the various operational and developmental UFS applications as well as tangential developments in tracking the erodibility of soils and the dust record.

Dr. Barry Baker is a physical research scientist at NOAA Air Resources Laboratory. He received his PhD from the University of Maryland Baltimore County (UMBC) and later received a postdoctoral fellowship through the National Research Council. He is now also on the global steering committee for the WMO Sand and Dust Storm group.

Watch on Youtube
Presentation Slides

Presenter(s): Dr. Zhan Zhang, Dr. Xuejin Zhang
A new operational hurricane prediction system for NOAA: UFS-based Hurricane Analysis and Forecast System (HAFS)

Hurricane Analysis and Forecast System (HAFS) is a Unified Forecast System(UFS) based hurricane application developed in the UFS R2O project. It is an atmosphere-ocean-wave coupled Tropical Cyclone prediction system. It has five salient features: (1) storm-following telescopic moving nests, (2) high-resolution physics configured for TC application, (3) storm inner-core Data Assimilation (DA) with vortex initialization, (4) atmosphere-ocean-wave coupling framework, and (5) intensive hurricane observational platforms to support the storm-scale DA system as well as the physics calibrations and system verifications/validations. HAFS is scheduled to become operational on June 27, 2023 to replace NOAA’s existing operational TC forecast systems, HWRF and HMON.

In this talk, we will present detailed configuration of the first version of HAFS operational implementation and the scientific evaluation of the three-year (2020-2022) retrospective experiments. We will also outline the continuous development activities of HAFS in the Hurricane Integrated Application Team and the future plans for HAFS in the next year and beyond.

Dr. Zhan Zhang obtained his Ph.D degree in tropical meteorology from Florida State University. He has worked at various research and operational organizations. He is the Hurricane Modelling team lead at NCEP/EMC, and the co-lead of NOAA’s Unified Forecast System (UFS) research and operation (R2O) Hurricane Application Team. He specializes in Numerical Weather Prediction modeling, especially in tropical cyclone modeling. His research interests include hurricane ensemble forecasts, vortex initialization, hurricane inner-core data assimilation.

Dr. Xuejin Zhang is a Meteorologist employed in NOAA’s Atlantic Oceanographic and Meteorological Laboratory’s Hurricane Research Division. He studies tropical cyclone forecast and simulation, land-air-sea interaction, regional climate, and parallel computing during his more than two-decade career. His expertise is in numerical algorithms, atmospheric dynamics, model initialization, and microphysics parameterization. He is currently leading the NOAA’s Unified Forecast System (UFS) R2O Hurricane Application Team. He obtained his Ph.D. in NC State University.

Watch on Youtube
Presentation Slides

Presenter: Patrick C. Burke
The vision of Warn-on-Forecast is to enable extended lead time for individual local severe weather hazards through design of new storm-scale probabilistic tools. Namely, the Warn-on-Forecast System (WoFS) employs rapid data assimilation, rapid forecast updates, and visualization of output at 5-minute resolution to provide movie-like probabilistic forecasts of individual storms on the 0-6 hour, “watch-to-warning” scale. Operating a relocatable, 900-km squared regional domain since 2017, the experimental 3-km WoFS has been shown to improve forecasts in testbed environments and to influence real-world decision-making and public threat communication at NWS national and local offices. WoFS migrated to the Microsoft Azure cloud in 2022, unlocking the capability to run multiple domains and/or larger domains. The NSSL double moment microphysics scheme designed in the WoF group is now available in the UFS, and WoFS is slated for an operational transition to the NWS, as part of the UFS, by late decade. Developing tools and processes within the UFS to support 5-15 minute data assimilation cycles, 30-minute relaunch cadence, and low-latency generation of unique ensemble visualizations will pave the way for future applications on a multitude of small-scale analysis and forecast problems.

Patrick Burke worked as a National Weather Service forecaster, including 5 years as a Lead Forecaster at the Weather Prediction Center. In 2020, he brought his operational experience to the Warn-on-Forecast (WoF) research group at the National Severe Storms Laboratory; Patrick is the WoF Program Lead. He holds an MS in meteorology from the University of Oklahoma. Growing up in the severe weather culture of Oklahoma he has always been motivated to help people stay safe during severe storm and flash flood events. In his spare time, Patrick enjoys running and writing music.
Watch on youtube
Presentation Slides

Presenter: Dr. Ben Green
Subseasonal prediction remains a uniquely challenging problem because the timescale involved cannot take much advantage of the memory imparted by atmospheric initial conditions (leveraged for predictions shorter than ~2 weeks), or of the slowly-evolving boundary forcings (leveraged for predictions longer than ~3 months). Regardless, interest in subseasonal prediction has grown substantially over the past decade owing to the identification of so-called “forecasts of opportunity” and the potential benefits of these forecasts to numerous sectors of society. Recognizing the demand for subseasonal forecasts, NOAA has been developing a fully-coupled Earth system model under the Unified Forecast System (UFS) framework which will be responsible for global (ensemble) predictions at lead times of 0-35 days. The development has involved several prototype coupled UFS runs consisting of bimonthly initializations over a 7-year period for a total of 168 cases.

This webinar presents results from a study that leverages these existing baseline prototypes to isolate the impact of substituting (one-at-a-time) parameterizations for convection, microphysics, and boundary layer on 35-d forecasts. It is found that no particular configuration of atmospheric physics within the coupled UFS is uniformly better or worse for subseasonal prediction, based on several metrics including mean-state biases and skill scores for the Madden-Julian Oscillation, precipitation, and 2-m temperature. Importantly, the spatial patterns of many “first-order” biases (e.g., impact of convection on precipitation) are remarkably similar between the end of the first week and weeks 3-4, indicating that some subseasonal biases may be mitigated through tuning at shorter timescales. An additional convective parameterization test using a different baseline shows that attempting to generalize specific results within UFS may be misguided.

Dr. Ben Green is a Research Scientist at the University of Colorado CIRES and works at the NOAA Global Systems Laboratory. He has been affiliated with these institutions since 2015, when he joined the team as a postdoc. He is a member of the GSL subseasonal-to-seasonal branch and works on the coupled UFS.
His research interests focus on numerical modeling of the Earth system at multiple timescales, with a particular emphasis on parameterizations for air-sea interaction. Ben has his Bachelor, Master, and PhD degrees in meteorology from Penn State University.

Watch on Youtube
Presentation Slides

Presenter: Lucas Harris (NOAA/GFDL)
NOAA has unified around two world-leading modeling systems to carry out its minutes-to-millennia mission: the Unified Forecast System for weather applications and NWS operations, and the GFDL Seamless Modeling Suite for climate-focused prediction, projection, and understanding. These share several major community components, including the FV3 Dynamical Core and the MOM6 Ocean Model. The NOAA Research Global-Nest Initiative is a multi-institution partnership to develop multiscale "digital twins" of the earth system for medium-range and subseasonal prediction and for ultra-high resolution climate modeling. This initiative will take advantage of new capabilities in our foundational technologies (FV3, SHiELD, FMS, Pace, etc.) to create global-nested convective-scale models, global storm-resolving models, and experimental kilometer and sub-kilometer scale process models. Results on cloud-radiation-prediction interactions, hurricanes, severe weather, the Madden-Julian Oscillation, and more will be shown. Progress on public availability, technology transfer, simulation on GPUs, and on large-eddy scale modeling will also be described, as will how this effort is acting to improve and extend both the UFS and SMS.

Lucas Harris has been interested in the weather since he was five years old. He is the Deputy Division Leader of the Weather and Climate Dynamics Division at NOAA’s Geophysical Fluid Dynamics Laboratory. His research is focused on development and application of the GFDL Finite-Volume Cubed-Sphere Dynamical Core (FV3) to create new models for frontier weather and climate problems. He holds a PhD in Atmospheric Sciences and an MS in Applied Mathematics, both from the University of Washington, and previously worked at Princeton University and NCAR.

Watch on youtube
Presentation Slides

Presenter:  Tara Jensen is Project Manager II at NCAR/RAL
With the evolution of weather and climate prediction using the Unified Forecast System (UFS), verification and evaluation activities are critical for the success of Research to Operations (R2O) across the UFS community. The enhanced Model Evaluation Tools (MET) framework (METplus) is at the core of an expanding cross-section of the community evaluation activities.

The METplus system consists of several components for the computation of traditional statistics based on gridded forecasts and either a gridded analysis or point-based observations. The system also incorporates an analysis system for aggregating statistics and plotting graphical results. These tools are designed to be highly flexible to allow for quick adaption to meet additional evaluation and diagnostic needs. A suite of python wrappers have been implemented in METplus to facilitate a low-level automation and configuration of the system, and to enhance the pre-existing plotting capabilities.

This presentation will briefly highlight advances in the recent METplus coordinated 5.0.0 release. It will also address the challenges and successes of community contributions and established capability to facilitate the advancement of innovations through UFS stages and gates.

Tara Jensen is Project Manager II at NCAR/RAL and the Deputy for the NCAR node of the Developmental Testbed Center. For the past 8 years, she has been leading the DTC verification team in the development of the enhanced Model Evaluation Tools verification and diagnostics framework (METplus). This framework has its roots in Tropospheric weather and Tropical Cyclone verification but over the years has been extended to also specifically include metrics for applications such as Subseasonal to Seasonal, Atmospheric Composition, Marine and Cryosphere. Tara and her team have also been working with SWPC over the past three years to guide development of METplus for use in Space Weather model evaluation.
Watch on youtube
Presentation Slides