NWS Rolls Out Improved Short-Range Forecast Model After Decade of Testing

After more than a decade, the National Weather Service has officially adopted an experimental short-range weather model capable of providing more precise predictions under rapidly changing storm conditions.

The new model will provide scientists with a better understanding for thunderstorm and winter snowstorm threats on a local level.

"The High-Resolution Rapid Refresh (HRRR) model updates on an hourly basis," HRRR developer Stan Benjamin said. "HRRR provides more time-specific information using the latest atmospheric conditions."

Benjamin is the chief of the Earth Modeling Branch in the Global Systems Division of NOAA's Earth System Research Lab.

According to the University Corporation for Atmospheric Research (UCAR) website, the foundations for HRRR's inception came from multiple NOAA agencies through the Weather Research and Forecasting (WRF) model and NCAR-based version, the ARW.

Assisted by his colleagues at the National Center for Atmospheric Research (NCAR) and UCAR, Benjamin and his team continued to make improvements to the existing NCAR model, which was the basis for HRRR.

The HRRR will provide a better resolution and can simulate individual thunderstorms with grid points of about 2 miles apart, according to UCAR's website.

In addition, Benjamin said that during the winter, the model can help provide better forecasts for snowstorms.

After several years of experimental use, Benjamin and those who worked on developing the HRRR model are now able to provide NWS and the public with access to more detailed storm information, precise 15-hour forecasts that can be adapted quickly and provide a better understanding of evolving storms.

"That's very gratifying for us," he said, referring to the years of testing and efforts involved in moving the model out of the experimental stage.

The success and capabilities of HRRR, according to Benjamin, is a testament to not only the ability to collect radar data every hour and radar assimilation, but also to advances in technology.

"We can improve [the representation of] environments for summertime thunderstorms," he said. "The environment for thunderstorms is so critical in getting a better and more accurate forecast."

Through the HRRR, identifying the properties that give rise to rotating thunderstorms could also help advance scientists' understanding of tornado development and provide a better lead time in warning the public of potentially life-threatening hazards, he said.

"It's pretty close to real time; we have to wait for the model, but we get a 15-hour forecast," Benjamin added.

While the model currently takes about an hour or more to provide the 15-hour forecast, Benjamin said as technology improves, the time will also improve.

"I think there has been a lot of progress and I think it will continue," he said, adding that it is reasonable to think that the process could be reduced to 30 to 40 minutes in the future.