Saturday, March 24, 2018
NOAA’s Upcoming Weather Forecast Model Zeros In Earlier on Severe Weather

NOAA’s Upcoming Weather Forecast Model Zeros In Earlier on Severe Weather

Research behind the High-Resolution Rapid Refresh weather forecast model

When it comes to weather, the more you know and the sooner you know it, the safer and better prepared you can be.

 Later this year, NOAA’s National Weather Service will usher into daily operations a sophisticated model called the High-Resolution Rapid Refresh, or HRRR, that will update forecasts hourly over the entire lower 48 United States at extremely sharp resolution using the latest observations from a network of ground and satellite-based sensors, radars and aircraft.

The HRRR provides forecast information at a resolution four times finer than what is currently used in hourly updated NOAA models.  This improvement in resolution from 13 to three kilometers is like giving forecasters an aerial photograph in which each pixel represents a neighborhood instead of a city.

 “When a typical thunderstorm is about 10 to 20 kilometers across, contains both upward and downward air currents as well as other features that give clues to its potential to create dangerous weather, it’s important to be able to see what’s happening inside the storm,” said Stan Benjamin, a research meteorologist at NOAA’s Earth System Research Lab in Boulder, Colo., who leads the team that developed the HRRR.   “It’s a game-changer to go from 13 to three kilometers for model resolution.”

El Reno Tornado

El Reno Tornado

During the early evening of May 31, 2013, the widest tornado in recorded history occurred over rural areas of central Oklahoma touching down in El Reno. (Photo courtesy of Jeff Snyder)
HRRR can differentiate between rotating storms, which are more likely to produce large tornadoes, and non-rotating thunderstorms that are often less dangerous. The HRRR can also predict damaging straight-line windstorms called “derechoes” and provide key timing information about when storm severity is increasing and when it’s decreasing. It can provide a wealth of information to forecasters about other weather hazards such as narrow heavy snow bands found in winter storms and areas of low ceiling and visibility. This information is vital to safer and more economic planning for transportation, including decisions by pilots and air traffic managers trying to maneuver planes around hazardous conditions.

Tested and refined across the country for the past two years, including on a number of severe weather events, the HRRR has demonstrated its effectiveness. On June 29, 2012, the HRRR helped identify a major “derecho” severe wind event as it was forming in Illinois and forecast its rapid sweep through Indiana and Ohio during the day and into the Washington, D.C., area that evening. It was also instrumental in providing the public with early forecasts of the May 31, 2013 tornadoes in Oklahoma and flash flooding that day.

High Resolution forecasts Derecho

The High-Resolution Rapid Refresh forecast model was the only NOAA model to capture in advance the June 29, 2012 derecho that struck the Washington, DC region. Shown on the right is a video of the HRRR forecast beginning at 11am Eastern Time that day, made available by NOAA in the early afternoon. On the left is the actual radar of the storm, which shows that the HRRR forecast closely aligned with what occurred. (NOAA)
“The HRRR’s higher resolution, advanced physics and enhanced data processing, have made a noticeable improvement in forecasting smaller scale weather,” said Steve Zubrick, the Science and Operations Officer for the National Weather Service Baltimore-Washington Forecast Office. “It’s allowed us to better anticipate evolving weather from wind, fog, precipitation and thunderstorms on time scales less than 12 hours.”

Led by Benjamin, researchers at NOAA’s Earth System Research Lab began developing the HRRR five years ago in response to several important developments: widespread recognition that a higher resolution model would improve severe storm forecasts; rapidly improving computer power required to process weather data at the higher resolution; and improved access to national radar data. Researchers also developed detailed techniques to use observations from radar, commercial aircraft, satellites, weather balloons, and surface stations often near airports to allow the HRRR model to differentiate between various weather situations.

The Earth System Research Lab, Global Systems Division, has worked collaboratively with a robust team to refine the HRRR. Key players on the team are NOAA’s National Weather Service including its National Centers for Environmental Prediction, NOAA’s cooperative institute partners; the Cooperative Institute for Research in Environmental Science at the University of Colorado at Boulder and the Cooperative Institute for Research in the Atmosphere at Colorado State University, Fort Collins, the Federal Aviation Administration, the National Center for Atmospheric Research, and the Department of Energy.

For more information, please contact Monica Allen, director of public affairs @NOAA Research, 301-734-1123 or



Previous Article Amazonian drought conditions add carbon dioxide to the atmosphere
Next Article Mapping climate change in the oceans