Raw model output on the native 0.25° grid — no MOS, no bias correction, no post-processing. Scores understate operational forecast quality. Read the fairness statement →

CastCheck methodology v0.2 · data through 2026-08-30

San Francisco Intl · GraphCast (IFS init) · lead day 7

KSFO · graphcast_ifs · lead 7

Target day = initialization date + 7 days. This page is a permanent link: the numbers change as history accumulates, the address and the methodology version do not.

Scores use the latest model version segment, GRAP_v100, starting 2026-01-01 — days before that ran a different version of the model and are excluded rather than silently pooled. Of the scored days, 1 carry a QC flag on the observation.

JSON for this page CSV of the daily errors Cite this page

Every number for this combination

All four windows × both initializations × both interpolations × both variables. Errors are computed in °C and shown in °F; CI is a 95 % moving-block bootstrap interval. MAE debiased and skill, debiased remove the constant offset of the window first, so a station that the model always reads a degree too warm is not scored as if it had no skill — compare the two skill columns to see how much of the error is a fixed offset. n QC next to n counts days whose observation carries a QC flag.

VariableInitMethod Windown MAE °F MAE debiased Bias °F RMSE °F ±1 °F±2 °F±3 °F Skill Skill, debiased Period
tmax00Zbilinear30d23 1 QC5.95 [4.88, 6.97]2.56-5.74 [-6.95, -4.33]6.54 [5.31, 7.57]0% [0%, 0%]0%17%-0.65+0.272026-08-08 → 2026-08-30
tmax00Zbilinear90d23 1 QC5.95 [4.62, 7.30]2.56-5.74 [-7.30, -3.94]6.54 [5.11, 7.86]0% [0%, 0%]0%17%-0.65+0.272026-08-08 → 2026-08-30
tmax00Zbilinear365d29 1 QC5.12 [3.14, 6.66]2.65-4.91 [-6.63, -2.73]5.95 [3.86, 7.31]10% [0%, 33%]10%28%-0.41+0.232026-01-08 → 2026-08-30
tmax00Zbilinearall29 1 QC5.12 [2.79, 6.70]2.65-4.91 [-6.64, -2.46]5.95 [3.49, 7.33]10% [0%, 36%]10%28%-0.41+0.232026-01-08 → 2026-08-30
tmax00Znearest30d23 1 QC2.89 [2.06, 3.78]2.89+0.86 [-0.52, 2.87]3.67 [2.58, 4.83]22% [3%, 40%]43%61%+0.20+0.182026-08-08 → 2026-08-30
tmax00Znearest90d23 1 QC2.89 [1.93, 4.02]2.89+0.86 [-0.47, 3.42]3.67 [2.45, 5.08]22% [0%, 40%]43%61%+0.20+0.182026-08-08 → 2026-08-30
tmax00Znearest365d29 1 QC2.75 [1.90, 3.59]2.75+0.36 [-1.06, 2.14]3.50 [2.42, 4.56]24% [0%, 40%]45%62%+0.24+0.202026-01-08 → 2026-08-30
tmax00Znearestall29 1 QC2.75 [1.89, 3.55]2.75+0.36 [-1.08, 2.15]3.50 [2.42, 4.54]24% [6%, 40%]45%62%+0.24+0.202026-01-08 → 2026-08-30
tmax12Zbilinear30d23 1 QC5.62 [4.78, 6.46]2.10-5.39 [-6.42, -4.14]6.12 [5.09, 7.16]0% [0%, 0%]0%17%-0.56+0.402026-08-08 → 2026-08-30
tmax12Zbilinear90d23 1 QC5.62 [4.54, 6.65]2.10-5.39 [-6.65, -3.76]6.12 [4.89, 7.45]0% [0%, 0%]0%17%-0.56+0.402026-08-08 → 2026-08-30
tmax12Zbilinear365d28 1 QC4.97 [3.23, 6.22]2.22-4.74 [-6.22, -2.66]5.63 [3.78, 6.88]7% [0%, 25%]7%29%-0.33+0.372026-01-08 → 2026-08-30
tmax12Zbilinearall28 1 QC4.97 [2.88, 6.18]2.22-4.74 [-6.15, -2.44]5.63 [3.40, 6.83]7% [0%, 28%]7%29%-0.33+0.372026-01-08 → 2026-08-30
tmax12Znearest30d23 1 QC2.51 [1.38, 4.05]2.44+1.53 [0.25, 3.53]3.54 [1.86, 5.05]39% [16%, 62%]57%65%+0.30+0.302026-08-08 → 2026-08-30
tmax12Znearest90d23 1 QC2.51 [1.30, 4.48]2.44+1.53 [0.19, 4.24]3.54 [1.75, 5.47]39% [11%, 63%]57%65%+0.30+0.302026-08-08 → 2026-08-30
tmax12Znearest365d28 1 QC2.39 [1.33, 3.76]2.36+0.99 [-0.58, 2.84]3.33 [1.70, 4.78]36% [15%, 58%]57%68%+0.36+0.332026-01-08 → 2026-08-30
tmax12Znearestall28 1 QC2.39 [1.35, 3.74]2.36+0.99 [-0.56, 2.87]3.33 [1.80, 4.75]36% [13%, 58%]57%68%+0.36+0.332026-01-08 → 2026-08-30
tmin00Zbilinear30d23 1 QC1.81 [1.20, 2.55]1.32-1.26 [-2.45, -0.12]2.10 [1.38, 2.78]26% [7%, 45%]52%78%-0.61-0.172026-08-08 → 2026-08-30
tmin00Zbilinear90d23 1 QC1.81 [1.18, 2.84]1.32-1.26 [-2.71, 0.22]2.10 [1.36, 2.99]26% [0%, 46%]52%78%-0.61-0.172026-08-08 → 2026-08-30
tmin00Zbilinear365d29 1 QC1.98 [1.27, 2.81]1.87-0.45 [-2.11, 1.46]2.33 [1.47, 3.11]24% [7%, 43%]48%72%+0.32+0.452026-01-08 → 2026-08-30
tmin00Zbilinearall29 1 QC1.98 [1.30, 2.84]1.87-0.45 [-2.00, 1.67]2.33 [1.47, 3.09]24% [7%, 43%]48%72%+0.32+0.452026-01-08 → 2026-08-30
tmin00Znearest30d23 1 QC1.76 [1.51, 2.11]1.16-1.24 [-1.87, -0.42]1.99 [1.78, 2.28]22% [9%, 30%]52%91%-0.55-0.022026-08-08 → 2026-08-30
tmin00Znearest90d23 1 QC1.76 [1.49, 2.23]1.16-1.24 [-1.94, -0.16]1.99 [1.78, 2.38]22% [0%, 30%]52%91%-0.55-0.022026-08-08 → 2026-08-30
tmin00Znearest365d29 1 QC1.70 [1.41, 2.09]1.10-1.29 [-1.87, -0.48]1.95 [1.71, 2.26]24% [8%, 38%]55%93%+0.42+0.682026-01-08 → 2026-08-30
tmin00Znearestall29 1 QC1.70 [1.41, 2.04]1.10-1.29 [-1.82, -0.50]1.95 [1.69, 2.20]24% [10%, 38%]55%93%+0.42+0.682026-01-08 → 2026-08-30
tmin12Zbilinear30d23 1 QC1.86 [1.19, 2.64]1.23-1.37 [-2.55, -0.28]2.12 [1.38, 2.78]17% [0%, 35%]61%78%-0.64-0.092026-08-08 → 2026-08-30
tmin12Zbilinear90d23 1 QC1.86 [1.17, 2.86]1.23-1.37 [-2.76, -0.02]2.12 [1.36, 2.97]17% [0%, 36%]61%78%-0.64-0.092026-08-08 → 2026-08-30
tmin12Zbilinear365d28 1 QC1.91 [1.22, 2.64]1.64-0.77 [-2.30, 0.93]2.20 [1.46, 2.82]18% [0%, 33%]57%79%+0.30+0.492026-01-08 → 2026-08-30
tmin12Zbilinearall28 1 QC1.91 [1.21, 2.67]1.64-0.77 [-2.20, 1.05]2.20 [1.46, 2.85]18% [0%, 34%]57%79%+0.30+0.492026-01-08 → 2026-08-30
tmin12Znearest30d23 1 QC1.98 [1.67, 2.42]1.11-1.48 [-2.15, -0.65]2.22 [1.87, 2.69]17% [7%, 25%]57%83%-0.75+0.012026-08-08 → 2026-08-30
tmin12Znearest90d23 1 QC1.98 [1.69, 2.54]1.11-1.48 [-2.25, -0.41]2.22 [1.88, 2.81]17% [0%, 26%]57%83%-0.75+0.012026-08-08 → 2026-08-30
tmin12Znearest365d28 1 QC1.97 [1.68, 2.38]1.10-1.53 [-2.13, -0.70]2.21 [1.88, 2.63]18% [7%, 28%]54%82%+0.27+0.652026-01-08 → 2026-08-30
tmin12Znearestall28 1 QC1.97 [1.65, 2.37]1.10-1.53 [-2.10, -0.77]2.21 [1.85, 2.66]18% [9%, 28%]54%82%+0.27+0.652026-01-08 → 2026-08-30

+ model too warm · − model too cold · bias interval includes zero · ★ lowest MAE · = not distinguishable from the leader · ▼ significantly worse · n < 30 greyed and unranked · CI is a 95 % moving-block bootstrap interval.

Daily error, last 90 days

Daily forecast minus observed error in °F for GraphCast (IFS init) at KSFO, lead day 7, 00Z bilinear tmax+20+100-10-202026-08-08: -2.64 °F2026-08-09: -5.29 °F2026-08-10: -9.91 °F2026-08-11: -11.35 °F2026-08-12: -8.34 °F2026-08-13: -9.4 °F2026-08-14: -2.68 °F2026-08-15: -5.09 °F2026-08-16: -7.26 °F2026-08-17: -9.8 °F2026-08-18: -4.42 °F2026-08-19: -5.29 °F2026-08-20: -4.58 °F2026-08-21: -4.02 °F2026-08-22: -3.61 °F2026-08-23: -7.69 °F2026-08-24: -8.97 °F2026-08-25: -2.07 °F2026-08-26: -8.39 °F2026-08-27: -3.42 °F2026-08-28: +2.48 °F2026-08-29: -5.5 °F2026-08-30: -4.69 °F2026-08-082026-08-30°F, forecast − observed
Forecast − observed, °F, 00Z bilinear daily maximum. The dashed line is zero; the thick line is a centred 7-day mean, drawn only where the whole window is present. Gaps are days with no score, never interpolation. Hover a point for its date and value.
The same series as a table (23 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-4.69
2026-08-29-5.50
2026-08-28+2.48
2026-08-27-3.42
2026-08-26-8.39
2026-08-25-2.07
2026-08-24-8.97
2026-08-23-7.69
2026-08-22-3.61
2026-08-21-4.02
2026-08-20-4.58
2026-08-19-5.29
2026-08-18-4.42
2026-08-17-9.80
2026-08-16-7.26
2026-08-15-5.09
2026-08-14-2.68
2026-08-13-9.40
2026-08-12-8.34
2026-08-11-11.35
2026-08-10-9.91
2026-08-09-5.29
2026-08-08-2.64

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KSFO, lead day 7-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 0 days-12 to -10 °F: 1 day-10 to -8 °F: 6 days-8 to -6 °F: 2 days-6 to -4 °F: 9 days-4 to -2 °F: 7 days-2 to 0 °F: 1 day0 to 2 °F: 2 days2 to 4 °F: 1 day4 to 6 °F: 0 days6 to 8 °F: 0 days8 to 10 °F: 0 days10 to 12 °F: 0 days12 to 14 °F: 0 days14 to 16 °F: 0 days16 to 18 °F: 0 days18 to 20 °F: 0 daysdashed: P50 |e| 4.69 · P90 |e| 9.48 °F-200+20 °Fn = 29 days
All scored days for 00Z bilinear daily maximum, 20 equal bins. The dashed marks are the 50th and 90th percentiles of the absolute error, mirrored on both sides. Hover a bar for its bin and count.
Bin counts as a table
From °FTo °FDays
-20.00-18.000
-18.00-16.000
-16.00-14.000
-14.00-12.000
-12.00-10.001
-10.00-8.006
-8.00-6.002
-6.00-4.009
-4.00-2.007
-2.00+0.001
+0.00+2.002
+2.00+4.001
+4.00+6.000
+6.00+8.000
+8.00+10.000
+10.00+12.000
+12.00+14.000
+14.00+16.000
+16.00+18.000
+18.00+20.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08235.95 -5.74
2026-0161.94 -1.75

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 29
GraphCast (IFS init)2026-01-08 → 2026-08-30 28
GraphCast (IFS init)2026-01-08 → 2026-08-30 28

Compared with other models

Paired bootstrap on the days both models cover (90d, 00Z, bilinear, daily maximum). Negative means GraphCast (IFS init) has the lower MAE. "Distinguishable" means the 95 % interval excludes zero.

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 23 +0.61 [-1.14, 1.59] no
NCEP GFS 23 +1.09 [-0.44, 2.50] no
ECMWF IFS HRES 18 -0.30 [-1.78, 1.51] no
Persistence (baseline) 23 +2.34 [-0.14, 4.04] no

Cite this page

ShortCastCheck, KSFO graphcast_ifs lead 7, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KSFO/model/graphcast_ifs/lead/7/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KSFO (San Francisco Intl), model graphcast_ifs, lead day 7. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KSFO/model/graphcast_ifs/lead/7/

Accessed2026-08-31 (data through 2026-08-30; page built 2026-08-31T03:52:06+00:00)

JSON CSV all scores methodology v0.2 licence