Skip to content

NOAA Research Global-Nest Initiative

The NOAA Research Global-Nest Initiative aims to develop convective-scale digital twins of the earth system for prediction, projection, and understanding of extreme weather events, and to create actionable information at all time scales.


NEWS

November 2023: The annual report for the first year’s activities (2022-2023) of the Global-Nest Initiative has been released. 

April 2023: The Global-Nest Initiative was the subject for this month’s UFS Webinar. You can watch it on YouTube here, or see the slides here.


Global-Nest Initiative: Smagorinsky’s Dream

The Global-Nest Initiative takes new technologies developed at GFDL and partners to create new model applications and products to better simulate and predict extreme weather events, their impacts, and their role within the broader earth system, within a holistically-designed unified prediction modeling system: a long-standing dream of model development going back to the earliest days of Numerical Weather Prediction. This is done with the powerful System for High-resolution prediction on Earth-to-Local Domains (SHiELD), an FV3-based weather prediction system.

GFDL SHiELD incorporates new technologies and paradigms for model development:

SHiELD forms the weather-to-subseasonal component of the GFDL Seamless Modeling Suite, and uses many of the same components as the Unified Forecast System (UFS), including FV3, FMS, GSI data assimilation system, and atmospheric physical parameterizations. The Global-Nest Initiative is centered in the FV3 Team and collaborates closely with UFS and GFDL Climate Model development, sharing many of the same open-source codes. This initiative also provides the earth system component of NOAA’s digital twin efforts for earth observations, ecosystems, fisheries, and human society.

The principal goals of the Global-Nest Initiative is to produce three seamlessly-integrated SHiELD configurations as Multiscale Earth System Digital Twins:

  • Global SHiELD (6-km deterministic, 16-km ensemble)
  • Global-to-regional nested runs (2-km deterministic, 5-km ensemble) over the Contiguous United States (CONUS) and tropical North Atlantic Ocean, with global SHiELD as its parent
  • X-SHiELD Global storm-resolving model (GSRM; 3-km)

and use them to create three principal products:

  • Deterministic 10-day high-resolution predictions 
  • Ensemble of 30+ day subseasonal predictions
  • Climate change simulations forced by projected ocean temperatures, land-use changes, and/or enhanced greenhouse gases

Our principal focus is on extreme weather events worldwide and specifically in the United States, especially:

  • Landfalling hurricanes
  • Atmospheric rivers
  • Tornado outbreaks and severe wind and hailstorms
  • “Bomb” winter storms
  • Wet and dry (“hydroclimate”) extremes
  • Flooding and extreme winds in mountainous regions

and on how they are affected by, and in turn, impact the larger-scale atmospheric circulations, external climate forcing, and human society. We aim to produce actionable local-scale information on both weather and climate timescales for stakeholders, scientists, and other downstream users, which includes real-time deterministic and ensemble products, climate-risk information, and metrics for the reliability and skill of predictions.

The Global-Nest Initiative also aims to explore even newer model configurations, including multiple-level and moving nests to reach kilometer-scale and sub-kilometer scale simulations, small enough to resolve urban areas, steep terrain, and tornado vortices.

The NOAA Research Global-Nest Initiative is a congressionally-mandated, base-funded project in the NOAA Climate Portfolio, aligned with the NOAA Research strategic priorities of “Make Forecasts Better” and “Drive Innovative Science”. This initiative is a collaboration between

  • NOAA Geophysical Fluid Dynamics Laboratory;
  • NOAA Atlantic Meteorological and Oceanic Laboratory;
  • NOAA National Severe Storms Laboratory;
  • Princeton University’s Program on Atmospheric and Oceanic Sciences;
  • NOAA Global Systems Laboratory;
  • the University Corporation for Atmospheric Research;
  • NOAA Environmental Modeling Center;
  • and the Allen Institute for Artificial Intelligence.