Radar Forecast
Forecast model
RRFS
The Rapid Refresh Forecast System.
RRFS is NOAA's next convection-allowing model, and the successor to HRRR. It is what takes the app's timeline past 48 hours and out to 84.
RRFS is in preview. It is planned to become operational on 14 October 2026. Until then its output is experimental, published on a preliminary basis, and can change or stop without notice. Weigh it accordingly — and weigh it less the further out you look.
At a glance
| Run by | NOAA / NCEP |
|---|---|
| Grid spacing | 3 km |
| Domain | North America |
| Dynamical core | FV3 |
| Initialised | Every hour |
| Forecast length | 84 hours from the 00, 06, 12 and 18 Z cycles |
| Status | Preview; operational planned 14 October 2026 |
| In the app | Timeline hours 49–84 |
One system instead of four.
NOAA currently runs several overlapping short-range models, each with its own grid, its own schedule and its own quirks. RRFS is the consolidation: a single hourly-updated convection-allowing system, on one 3 km grid, covering all of North America rather than only the lower 48.
It is built on the FV3 dynamical core — the same finite-volume core NOAA adopted for its global model — where HRRR is built on WRF. That change is the reason RRFS is a new system rather than another HRRR version.
Why it earns the last 36 hours.
HRRR's synoptic runs stop at 48 hours. RRFS reaches 84 from the same 00, 06, 12 and 18 Z cycles, which is the difference between planning tomorrow and planning the day after.
The app uses that and nothing more. Hours 1–48 come from HRRR; only hours 49–84 come from RRFS. Where both models offer the same hour, HRRR wins — it is the operational one, and swapping models mid-timeline would change the look of the imagery for no gain in accuracy.
The two share a grid, which is what makes the joint seamless rather than a visible step in the shading.
When frames appear.
RRFS publishes its 19–84 hour frames roughly 3 hours 25 minutes after the cycle time — noticeably later than HRRR's 1 hour 45, which is what you would expect from a run nearly twice as long over a larger domain.
If a cycle has not posted yet, the hours it would have filled stay empty rather than falling back to something older and wrong. The slider keeps its full length; the unavailable detents simply carry no image.
How far to trust hour 80.
Not very far, and that is not a criticism of the model. A convection-allowing forecast three days out will not tell you where a storm will be. What it will tell you is whether a system is expected to be over your route at all, and roughly when — which is enough to decide whether to look harder at that day.
Use the far end of the timeline to choose which day to fly. Use the near end to choose the route.
Compared with what actually happened
The site publishes forecast frames against the observed radar for the same valid time, so you can judge how the models did rather than take anyone's word for it.
A forecast is wrong in the details by definition, and a preview model doubly so. RRFS output in this app is experimental. Radar Forecast is a planning aid, not a certified EFB. Always get an official preflight briefing.