WEATHERMAPS.AI · THE AI MODEL DESK OF THE FORECAST NETWORK FORECAST EAGLE FORECAST EURO WEATHER.CITY WEATHER.FOOTBALL
weathermaps.ai
The Board · every model, every run, on one page

The atmosphere, learned and solved, side by side

weathermaps.ai tracks the machine-learning weather models against the physics models they are measured against. One model's map carries almost no information, because every model draws a plausible Tuesday. What carries information is where two of them, run at the same hour, stop agreeing.

On the Board 5 of 14 tracked 2 learned · 3 physics
Newest cycle 16 Sep 18Z 7 h ago
Feeds on time 5 / 5 Status page →
Disagreement · day 5 44 NOTABLE 90th percentile of 602 cities · 16 Sep 12Z run

Most contested cities

Where AIFS, HRES, WeatherNext 2 and the ensemble mean disagree most, measured against how much the ensemble itself expects. From the 16 Sep 12Z run. All 602 cities →

50 = as much as the ensemble expects
City DI Driven by Highs, °C to Mon 21 12Z
LubumbashiCongo (Kinshasa) 73 temperature 23322531 spread 9°
BrasíliaBrazil 60 temperature 31312931 spread 2°
EdmontonCanada 60 temperature 24212322 spread 3°
MemphisUnited States 60 temperature 36383438 spread 4°
AsunciónParaguay 59 pressure 35313332 spread 4°
JiningChina 58 temperature 24292529 spread 5°
CuritibaBrazil 57 pressure 21252123 spread 4°
Cape TownSouth Africa 56 temperature 23272125 spread 6°
LusakaZambia 56 temperature 22302430 spread 8°
LudhianaIndia 55 temperature 35323132 spread 4°
ShangqiuChina 55 temperature 22282328 spread 6°
XuzhouChina 54 temperature 24292429 spread 5°

Runs coming in

Each model's newest run and how far through its forecast the rendered maps have reached. Unfinished runs come first.

Model Board

Sorted by family, then by newest run. Click a column to sort it, and a model to open it in the explorer.

updated 17 Sep 01:03Z
Model Status Newest run Horizon Fields Grid Rendered by
AIFS SingleECMWF · Deterministic On time 16 Sep 12Z13 h ago f360 11 0.25° forecasteuro.com
WeatherNext 2Google DeepMind · Ensemble (64 members) On timeExperimental 16 Sep 12Z13 h ago f360 4 0.25° forecasteuro.com
IFS HRESECMWF · Deterministic On time 16 Sep 12Z13 h ago f360 24 0.1° forecasteuro.com
IFS ENSECMWF · Ensemble On time 16 Sep 12Z13 h ago f360 72 0.25° forecasteuro.com
GFSNOAA NCEP · Deterministic On time 16 Sep 18Z7 h ago f240 8 0.117° forecasteagle.com

Open a model

Every explorer uses the same colours, places and timeline, so switching models changes only the forecast.

Why this site exists

The baseline moved

For fifty years, better forecasts meant a better dynamical core and a bigger machine. Then a model that had never been told what a pressure gradient is started beating one built out of them. That is worth watching closely rather than through a press release.

Difference is the signal

One model's map tells you almost nothing. What carries information is where two of them, run at the same hour, stop agreeing, so this site is built around comparing models rather than around any one of them.

Coarse, and honest about it

The global learned models here run at a quarter degree, about 28 km. Where detail appears at a finer scale than the grid, it was generated rather than forecast, and the page says which.