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 5

KSFO · graphcast_ifs · lead 5

Target day = initialization date + 5 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
tmax00Zbilinear30d25 1 QC5.56 [4.78, 6.32]2.36-5.29 [-6.14, -4.43]6.10 [5.33, 6.76]0% [0%, 0%]8%20%-0.31+0.422026-08-06 → 2026-08-30
tmax00Zbilinear90d25 1 QC5.56 [4.57, 6.37]2.36-5.29 [-6.22, -4.26]6.10 [5.12, 6.80]0% [0%, 0%]8%20%-0.31+0.422026-08-06 → 2026-08-30
tmax00Zbilinear365d31 1 QC4.89 [3.31, 5.98]2.49-4.65 [-5.78, -3.00]5.59 [4.00, 6.49]6% [0%, 23%]13%32%-0.25+0.302026-01-06 → 2026-08-30
tmax00Zbilinearall31 1 QC4.89 [3.14, 6.04]2.49-4.65 [-5.85, -2.86]5.59 [3.87, 6.54]6% [0%, 25%]13%32%-0.25+0.302026-01-06 → 2026-08-30
tmax00Znearest30d25 1 QC2.79 [1.62, 4.48]2.80+1.63 [0.31, 3.56]4.15 [2.21, 5.96]28% [12%, 43%]64%68%+0.34+0.312026-08-06 → 2026-08-30
tmax00Znearest90d25 1 QC2.79 [1.58, 4.88]2.80+1.63 [0.26, 4.25]4.15 [2.23, 6.26]28% [9%, 44%]64%68%+0.34+0.312026-08-06 → 2026-08-30
tmax00Znearest365d31 1 QC2.62 [1.55, 4.11]2.63+1.00 [-0.70, 2.95]3.85 [2.08, 5.57]29% [14%, 45%]61%71%+0.33+0.262026-01-06 → 2026-08-30
tmax00Znearestall31 1 QC2.62 [1.57, 4.02]2.63+1.00 [-0.68, 3.03]3.85 [2.08, 5.48]29% [14%, 44%]61%71%+0.33+0.262026-01-06 → 2026-08-30
tmax12Zbilinear30d25 1 QC5.46 [4.56, 6.33]2.40-5.35 [-6.23, -4.45]6.08 [5.29, 6.73]8% [0%, 16%]12%16%-0.29+0.412026-08-06 → 2026-08-30
tmax12Zbilinear90d25 1 QC5.46 [4.39, 6.46]2.40-5.35 [-6.39, -4.32]6.08 [5.08, 6.80]8% [0%, 17%]12%16%-0.29+0.412026-08-06 → 2026-08-30
tmax12Zbilinear365d30 1 QC4.94 [3.46, 6.05]2.47-4.86 [-6.02, -3.36]5.65 [4.14, 6.52]10% [0%, 20%]13%27%-0.25+0.322026-01-06 → 2026-08-30
tmax12Zbilinearall30 1 QC4.94 [3.33, 6.10]2.47-4.86 [-6.05, -3.32]5.65 [4.02, 6.58]10% [0%, 21%]13%27%-0.25+0.322026-01-06 → 2026-08-30
tmax12Znearest30d25 1 QC2.70 [1.52, 4.26]2.89+1.59 [0.24, 3.40]3.92 [2.29, 5.44]40% [17%, 64%]56%68%+0.36+0.292026-08-06 → 2026-08-30
tmax12Znearest90d25 1 QC2.70 [1.48, 4.53]2.89+1.59 [0.28, 3.93]3.92 [2.36, 5.71]40% [14%, 65%]56%68%+0.36+0.292026-08-06 → 2026-08-30
tmax12Znearest365d30 1 QC2.59 [1.54, 4.02]2.78+0.99 [-0.72, 2.89]3.70 [2.23, 5.15]37% [16%, 60%]53%73%+0.35+0.242026-01-06 → 2026-08-30
tmax12Znearestall30 1 QC2.59 [1.49, 4.02]2.78+0.99 [-0.87, 3.03]3.70 [2.20, 5.14]37% [14%, 61%]53%73%+0.35+0.242026-01-06 → 2026-08-30
tmin00Zbilinear30d25 1 QC1.47 [1.05, 1.98]1.17-0.91 [-1.93, 0.04]1.74 [1.22, 2.23]36% [16%, 57%]72%92%-0.31-0.022026-08-06 → 2026-08-30
tmin00Zbilinear90d25 1 QC1.47 [0.98, 2.11]1.17-0.91 [-2.07, 0.22]1.74 [1.16, 2.36]36% [12%, 59%]72%92%-0.31-0.022026-08-06 → 2026-08-30
tmin00Zbilinear365d31 1 QC1.53 [1.09, 2.00]1.41-0.62 [-1.82, 0.48]1.81 [1.28, 2.25]32% [14%, 54%]71%90%+0.32+0.452026-01-06 → 2026-08-30
tmin00Zbilinearall31 1 QC1.53 [1.09, 2.01]1.41-0.62 [-1.67, 0.52]1.81 [1.29, 2.28]32% [12%, 55%]71%90%+0.32+0.452026-01-06 → 2026-08-30
tmin00Znearest30d25 1 QC1.35 [1.13, 1.64]1.01-1.06 [-1.56, -0.55]1.59 [1.36, 1.84]44% [26%, 59%]76%96%-0.20+0.122026-08-06 → 2026-08-30
tmin00Znearest90d25 1 QC1.35 [1.10, 1.69]1.01-1.06 [-1.67, -0.51]1.59 [1.35, 1.86]44% [22%, 59%]76%96%-0.20+0.122026-08-06 → 2026-08-30
tmin00Znearest365d31 1 QC1.61 [1.23, 2.36]1.09-1.38 [-2.32, -0.74]2.00 [1.48, 2.82]35% [12%, 53%]68%94%+0.28+0.582026-01-06 → 2026-08-30
tmin00Znearestall31 1 QC1.61 [1.21, 2.37]1.09-1.38 [-2.35, -0.74]2.00 [1.46, 2.85]35% [8%, 55%]68%94%+0.28+0.582026-01-06 → 2026-08-30
tmin12Zbilinear30d25 1 QC1.46 [1.02, 2.02]1.23-0.96 [-1.96, -0.05]1.79 [1.25, 2.30]44% [22%, 65%]68%88%-0.30-0.072026-08-06 → 2026-08-30
tmin12Zbilinear90d25 1 QC1.46 [0.94, 2.11]1.23-0.96 [-2.08, 0.08]1.79 [1.19, 2.39]44% [20%, 68%]68%88%-0.30-0.072026-08-06 → 2026-08-30
tmin12Zbilinear365d30 1 QC1.43 [1.00, 1.91]1.31-0.75 [-1.81, 0.24]1.75 [1.20, 2.20]43% [24%, 65%]70%90%+0.29+0.432026-01-06 → 2026-08-30
tmin12Zbilinearall30 1 QC1.43 [1.01, 1.92]1.31-0.75 [-1.78, 0.27]1.75 [1.22, 2.22]43% [23%, 63%]70%90%+0.29+0.432026-01-06 → 2026-08-30
tmin12Znearest30d25 1 QC1.53 [1.31, 1.82]1.05-1.17 [-1.71, -0.61]1.77 [1.56, 2.04]32% [13%, 48%]72%92%-0.36+0.082026-08-06 → 2026-08-30
tmin12Znearest90d25 1 QC1.53 [1.30, 1.87]1.05-1.17 [-1.82, -0.59]1.77 [1.54, 2.09]32% [10%, 50%]72%92%-0.36+0.082026-08-06 → 2026-08-30
tmin12Znearest365d30 1 QC1.74 [1.40, 2.38]1.10-1.44 [-2.32, -0.79]2.05 [1.63, 2.73]27% [7%, 43%]67%90%+0.14+0.522026-01-06 → 2026-08-30
tmin12Znearestall30 1 QC1.74 [1.40, 2.47]1.10-1.44 [-2.38, -0.82]2.05 [1.65, 2.77]27% [4%, 43%]67%90%+0.14+0.522026-01-06 → 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 5, 00Z bilinear tmax+20+100-10-202026-08-06: -6.21 °F2026-08-07: -2.93 °F2026-08-08: +1.31 °F2026-08-09: -4.49 °F2026-08-10: -7.67 °F2026-08-11: -9.56 °F2026-08-12: -6.96 °F2026-08-13: -6.97 °F2026-08-14: -4.94 °F2026-08-15: -5.52 °F2026-08-16: -6.02 °F2026-08-17: -6.91 °F2026-08-18: -4.64 °F2026-08-19: -1.1 °F2026-08-20: -4.32 °F2026-08-21: -3.42 °F2026-08-22: -3.88 °F2026-08-23: -10.86 °F2026-08-24: -8.14 °F2026-08-25: -2.38 °F2026-08-26: -8.7 °F2026-08-27: -5.24 °F2026-08-28: +2.01 °F2026-08-29: -6.41 °F2026-08-30: -8.32 °F2026-08-062026-08-30°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 (25 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-8.32
2026-08-29-6.41
2026-08-28+2.01
2026-08-27-5.24
2026-08-26-8.70
2026-08-25-2.38
2026-08-24-8.14
2026-08-23-10.86
2026-08-22-3.88
2026-08-21-3.42
2026-08-20-4.32
2026-08-19-1.10
2026-08-18-4.64
2026-08-17-6.91
2026-08-16-6.02
2026-08-15-5.52
2026-08-14-4.94
2026-08-13-6.97
2026-08-12-6.96
2026-08-11-9.56
2026-08-10-7.67
2026-08-09-4.49
2026-08-08+1.31
2026-08-07-2.93
2026-08-06-6.21

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KSFO, lead day 5-20 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: 4 days-8 to -6 °F: 7 days-6 to -4 °F: 6 days-4 to -2 °F: 8 days-2 to 0 °F: 2 days0 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| 4.64 · P90 |e| 8.32 °F-200+20 °Fn = 31 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
-10.00-8.004
-8.00-6.007
-6.00-4.006
-4.00-2.008
-2.00+0.002
+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
2026-08255.56 -5.29
2026-0162.10 -1.99

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 31
GraphCast (IFS init)2026-01-06 → 2026-08-30 30
GraphCast (IFS init)2026-01-06 → 2026-08-30 30

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 25 +1.04 [0.65, 1.50] yes
NCEP GFS 25 +0.86 [-0.13, 1.60] no
ECMWF IFS HRES 18 -0.49 [-1.31, 1.44] no
Persistence (baseline) 25 +1.32 [-0.81, 2.71] no

Cite this page

ShortCastCheck, KSFO graphcast_ifs lead 5, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KSFO/model/graphcast_ifs/lead/5/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KSFO (San Francisco Intl), model graphcast_ifs, lead day 5. 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/5/

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