We heard the first boom around 20 minutes ago.
“Hear the thunder?” Helaine asked. It was tough to avoid.
Over on Facebook Diane Campbell posted, “Just heard the loudest rumble of thunder I ever heard in my life! =(( scared me!”
It was in the forecast – phew! Bad weather isn’t so bad if it’s expected. That’s my very limited perspective.
These thunderstorms forming overhead gave me a chance to look at a new tool: The High Resolution Rapid Refresh model (HRRR). It’s your tax dollars at work.
The HRRR is the only hourly updated, radar-initialized, storm-resolving model running at this time over the US (or internationally), to our best knowledge. As a higher-resolution nest inside the hourly-updated Rapid Refresh the HRRR is designed to provide rapidly updated model guidance on convective storms for
- air traffic management
- severe weather forecasting
- NOAA National Weather Service Warn-On Forecast
- eventually provide improved background fields for NWS Real-Time Mesoscale Analysis
- provide improved basis for other aviation hazard forecasts (e.g., wake vortex, ceiling, visibility, turbulence, inflight icing, terminal forecasts)
Allow me to cut through the mumbo jumbo. HRRR is very high resolution in both time (15 minute increment output) and space (3km resolution). It is a short range model that only goes out 15 hours from initialization. Since it takes a while to process you’re left with 12 or 13 usable hours.
Typically models give hints not full solutions. You can never take the exact placement of features as a given. I expect the HRRR will change that!
I am looking at the 1730z radar image and predicted radar image. They’re pretty close¹. Not 100%, but much better than what I’m used to seeing.
If HRRR is as good as I hope these showers and thunderstorms will be gone by 5:30 pm (give or take a little).
Progress is good.
¹ – Unfortunately, the maps are different domains and projections. That makes overlaying them impossible.