Was the forecast right for San Francisco on Tue Sep 22?
San Francisco, California · daily high temperature in °C · forecasts from each model's archived runs, verified against the Duboce Triangle (SFOC1) observations
14 days out, 4 models put San Francisco's Tue Sep 22 high anywhere from 19° to 24°; by the final call, the day before, the field had settled to a median of 19°. The day verified at 17° against the Duboce Triangle (SFOC1) observations.
ICON landed closest, off by 0.1°. GFS ran 3° warm.
What every model said, at every lead
Forecast highs for Tue Sep 22 as each model's run approached it, recorded from the last archived model run per day. NBM and The Spread's own blend are omitted because they average the other models.
| Lead | AIFS | AIGFS | ECMWF | GEM | GFS | ICON | UKMO | Range |
|---|---|---|---|---|---|---|---|---|
| 15d · Sep 7 | — | 18° | — | — | 22° | — | — | 18–22° |
| 14d · Sep 8 | 24° | 22° | 19° | — | 22° | — | — | 19–24° |
| 13d · Sep 9 | 22° | 29° | 23° | — | 27° | — | — | 22–29° |
| 12d · Sep 10 | 23° | 31° | 19° | — | 25° | — | — | 19–31° |
| 11d · Sep 11 | 24° | 21° | 18° | — | 21° | — | — | 18–24° |
| 10d · Sep 12 | 26° | 20° | 18° | — | 21° | — | — | 18–26° |
| 9d · Sep 13 | 20° | 24° | 17° | 18° | 26° | — | — | 17–26° |
| 8d · Sep 14 | 22° | 25° | 19° | 19° | 26° | — | — | 19–26° |
| 7d · Sep 15 | 19° | 31° | 17° | 23° | 26° | — | — | 17–31° |
| 6d · Sep 16 | 19° | 21° | 18° | 18° | 23° | 17° | 22° | 17–23° |
| 5d · Sep 17 | 19° | 21° | 18° | 20° | 23° | 18° | 19° | 18–23° |
| 4d · Sep 18 | 19° | 21° | 19° | 18° | 23° | 17° | 19° | 17–23° |
| 3d · Sep 19 | 19° | 20° | 18° | 18° | 21° | 17° | 16° | 16–21° |
| 2d · Sep 20 | 20° | 20° | 19° | 18° | 22° | 17° | 17° | 17–22° |
| 1d · Sep 21 | 20° | 19° | 18° | 18° | 21° | 17° | 19° | 17–21° |
| actual | 17° |
What one day can show
A score for one day isn't very meaningful; a useful ranking needs many days of data. See which model has been most accurate in San Francisco, or in every city The Spread scores.
Forecasts are what each model's archived runs said at the time they were recorded by The Spread; nothing is reconstructed after the fact. Truth is the Duboce Triangle (SFOC1) observations. The actual values are the ones in hand when this day was verified, usually 2 days after the date; a correction published after that is not carried into it. Archive depth varies by city; verified days exist only for dates the archive watched arrive.