Weather Research and Forecasting Model

Weather Research and Forecasting Model
Typhoon_Mawar_2005_computer_simulation.gif: Atmoz derivative work: Atmoz (talk) · CC BY-SA 3.0 · source
Name	Weather Research and Forecasting Model
Developer	National Center for Atmospheric Research; National Oceanic and Atmospheric Administration; U.S. Air Force
Initial release	2000s
Programming language	Fortran; Python (programming language) (utilities)
Operating system	Unix-like
Genre	Numerical weather prediction; atmospheric modeling

Weather Research and Forecasting Model

The Weather Research and Forecasting Model is a numerical simulation system used for mesoscale and regional atmospheric research and forecasting. It serves research communities across institutions such as the National Center for Atmospheric Research, National Oceanic and Atmospheric Administration, U.S. Air Force, and international agencies including the Met Office and Environment and Climate Change Canada. The system underpins studies tied to events like the Hurricane Katrina response and operations linked to FIFA World Cup venue planning.

Overview

The system provides a framework for mesoscale simulation, data assimilation, and scenario testing used by organizations such as the European Centre for Medium-Range Weather Forecasts, Japan Meteorological Agency, China Meteorological Administration, Indian Meteorological Department, and Australian Bureau of Meteorology. Research leveraging the model has been published in journals associated with the American Meteorological Society, Royal Meteorological Society, and Geophysical Research Letters and presented at conferences like the AGU Fall Meeting and AMS Annual Meeting. Major deployments intersect with programs from NASA missions, NOAA Hurricane Forecast Improvement Project, and military forecasting efforts by the United States Army and United States Navy.

Model Description and Components

The core solver uses nonhydrostatic, compressible equations and options tailored for physics parameterizations developed collaboratively by groups including NCAR and NOAA/NCEP. Key components include dynamic cores, boundary-layer schemes, microphysics packages, radiation modules, and surface-layer and land-surface models such as implementations linked to the Community Land Model and coupling frameworks used by DOE laboratories. The software interfaces with observational and reanalysis datasets from Global Forecast System, ERA5, NCEP Reanalysis, and satellite streams from GOES, MetOp, and Sentinel missions. Post-processing and visualization workflows integrate tools from Unidata, Python (programming language), and NCAR Command Language.

Development and History

Origins trace to collaborative projects among NCAR, NOAA, U.S. Air Force, and academic partners at institutions such as University of Oklahoma, Colorado State University, Penn State University, Massachusetts Institute of Technology, and University of Washington. Historical milestones align with community efforts represented at workshops hosted by UCAR and funded initiatives from agencies like NSF and NASA. Major version advances were driven by advances in high-performance computing from centers like NERSC and Oak Ridge National Laboratory, and code evolution paralleled developments in numerical methods described in literature from Courant–Friedrichs–Lewy-related studies and formulations used in many operational centers including NCEP.

Applications and Operational Use

Operationally, agencies including NOAA/NWS, Met Office, Environment and Climate Change Canada, Korean Meteorological Administration, and military forecasting units employ the system for convective-scale forecasting, wildfire smoke prediction during incidents like the 2018 Camp Fire, air quality modeling in coordination with the Environmental Protection Agency, and urban meteorology for events such as the Tokyo 2020 Olympics planning. It supports research into severe weather seen in Tornado Alley, tropical cyclone development like Hurricane Maria, orographic precipitation in ranges such as the Rocky Mountains and Himalayas, and coupling to hydrologic models used by agencies like the USGS for flood forecasting. Emergency management coordination has involved organizations such as the Federal Emergency Management Agency and international partners like the World Meteorological Organization.

Evaluation, Verification, and Performance

Verification studies compare output with observations from surface networks including ASOS, radiosonde campaigns coordinated by ARM (Atmospheric Radiation Measurement), radar networks like NEXRAD, and satellite retrievals from MODIS. Performance assessments are reported in venues such as Bulletin of the American Meteorological Society and use metrics developed by projects like the Verification of the Origins of Rotation in Tornadoes Experiment. High-performance implementations exploit architectures from Intel, NVIDIA, and supercomputing centers including Blue Waters and Fugaku, with scalability studies conducted alongside research labs including NOAA Geophysical Fluid Dynamics Laboratory.

Extensions, Coupling, and Community Contributions

The community maintains extensions and couplers linking to ocean models like HYCOM and Regional Ocean Modeling System, wave models such as WW3, air quality systems like CMAQ and CAMx, and data assimilation frameworks including GSI and 4D-Var efforts at centers like ECMWF. Open-source contributions come from universities and consortiums including UCAR, NCAR, NOAA ESRL, and international research groups at ETH Zurich and University of Reading. Training, tutorials, and code repositories are supported by workshops at AMS Summer Policy Colloquium and hands-on events sponsored by WCRP and national research facilities.

Category:Numerical weather prediction models