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 6

KSFO · graphcast_ifs · lead 6

Target day = initialization date + 6 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
tmax00Zbilinear30d24 1 QC5.60 [4.67, 6.46]2.44-5.34 [-6.39, -4.00]6.17 [5.16, 6.99]4% [0%, 12%]8%12%-0.49+0.322026-08-07 → 2026-08-30
tmax00Zbilinear90d24 1 QC5.60 [4.39, 6.53]2.44-5.34 [-6.52, -3.71]6.17 [4.80, 7.04]4% [0%, 12%]8%12%-0.49+0.322026-08-07 → 2026-08-30
tmax00Zbilinear365d30 1 QC4.95 [3.28, 6.21]2.48-4.72 [-6.17, -2.80]5.66 [3.96, 6.77]10% [0%, 26%]13%23%-0.36+0.282026-01-07 → 2026-08-30
tmax00Zbilinearall30 1 QC4.95 [3.11, 6.18]2.48-4.72 [-6.10, -2.76]5.66 [3.58, 6.77]10% [0%, 26%]13%23%-0.36+0.282026-01-07 → 2026-08-30
tmax00Znearest30d24 1 QC2.84 [2.02, 4.00]2.82+1.51 [0.10, 3.46]3.84 [2.65, 5.03]42% [24%, 58%]50%58%+0.24+0.212026-08-07 → 2026-08-30
tmax00Znearest90d24 1 QC2.84 [1.95, 4.35]2.82+1.51 [0.10, 4.27]3.84 [2.58, 5.27]42% [20%, 60%]50%58%+0.24+0.212026-08-07 → 2026-08-30
tmax00Znearest365d30 1 QC2.69 [1.91, 3.70]2.70+0.83 [-0.92, 2.85]3.60 [2.46, 4.71]40% [24%, 57%]47%60%+0.26+0.212026-01-07 → 2026-08-30
tmax00Znearestall30 1 QC2.69 [1.92, 3.65]2.70+0.83 [-0.91, 2.67]3.60 [2.44, 4.64]40% [23%, 57%]47%60%+0.26+0.212026-01-07 → 2026-08-30
tmax12Zbilinear30d24 1 QC5.46 [4.57, 6.20]2.35-5.17 [-6.06, -4.13]6.05 [5.30, 6.65]8% [0%, 18%]8%8%-0.46+0.342026-08-07 → 2026-08-30
tmax12Zbilinear90d24 1 QC5.46 [4.41, 6.29]2.35-5.17 [-6.18, -3.86]6.05 [5.09, 6.68]8% [0%, 20%]8%8%-0.46+0.342026-08-07 → 2026-08-30
tmax12Zbilinear365d29 1 QC4.81 [2.95, 5.98]2.56-4.56 [-5.82, -2.84]5.58 [3.93, 6.41]14% [0%, 33%]17%21%-0.30+0.252026-01-07 → 2026-08-30
tmax12Zbilinearall29 1 QC4.81 [2.81, 6.01]2.56-4.56 [-5.82, -2.54]5.58 [3.67, 6.48]14% [0%, 33%]17%21%-0.30+0.252026-01-07 → 2026-08-30
tmax12Znearest30d24 1 QC2.87 [1.80, 4.31]2.82+1.88 [0.64, 3.51]4.17 [2.50, 5.68]29% [16%, 42%]54%71%+0.24+0.212026-08-07 → 2026-08-30
tmax12Znearest90d24 1 QC2.87 [1.76, 4.68]2.82+1.88 [0.66, 4.05]4.17 [2.47, 6.10]29% [14%, 43%]54%71%+0.24+0.212026-08-07 → 2026-08-30
tmax12Znearest365d29 1 QC2.63 [1.58, 3.93]2.73+1.32 [-0.23, 3.03]3.88 [2.18, 5.38]31% [17%, 45%]55%72%+0.29+0.202026-01-07 → 2026-08-30
tmax12Znearestall29 1 QC2.63 [1.59, 3.88]2.73+1.32 [-0.40, 2.96]3.88 [2.06, 5.23]31% [17%, 45%]55%72%+0.29+0.202026-01-07 → 2026-08-30
tmin00Zbilinear30d24 1 QC1.55 [0.91, 2.31]1.21-1.10 [-2.23, -0.07]1.91 [1.20, 2.52]38% [12%, 63%]67%79%-0.61-0.242026-08-07 → 2026-08-30
tmin00Zbilinear90d24 1 QC1.55 [0.87, 2.49]1.21-1.10 [-2.44, 0.10]1.91 [1.14, 2.68]38% [10%, 65%]67%79%-0.61-0.242026-08-07 → 2026-08-30
tmin00Zbilinear365d30 1 QC1.67 [0.98, 2.38]1.52-0.63 [-2.01, 0.75]2.02 [1.33, 2.58]33% [11%, 58%]63%80%+0.33+0.482026-01-07 → 2026-08-30
tmin00Zbilinearall30 1 QC1.67 [1.04, 2.37]1.52-0.63 [-1.91, 0.74]2.02 [1.37, 2.60]33% [11%, 58%]63%80%+0.33+0.482026-01-07 → 2026-08-30
tmin00Znearest30d24 1 QC1.53 [1.32, 1.85]1.02-1.18 [-1.67, -0.59]1.82 [1.54, 2.19]38% [24%, 50%]62%96%-0.60-0.052026-08-07 → 2026-08-30
tmin00Znearest90d24 1 QC1.53 [1.34, 2.03]1.02-1.18 [-1.78, -0.52]1.82 [1.53, 2.34]38% [20%, 50%]62%96%-0.60-0.052026-08-07 → 2026-08-30
tmin00Znearest365d30 1 QC1.69 [1.35, 2.33]1.11-1.39 [-2.09, -0.75]2.09 [1.58, 2.85]37% [23%, 48%]60%90%+0.32+0.622026-01-07 → 2026-08-30
tmin00Znearestall30 1 QC1.69 [1.37, 2.29]1.11-1.39 [-2.14, -0.80]2.09 [1.60, 2.84]37% [25%, 47%]60%90%+0.32+0.622026-01-07 → 2026-08-30
tmin12Zbilinear30d24 1 QC1.61 [1.07, 2.26]1.19-1.18 [-2.22, -0.19]1.91 [1.27, 2.49]33% [12%, 54%]62%92%-0.67-0.222026-08-07 → 2026-08-30
tmin12Zbilinear90d24 1 QC1.61 [1.05, 2.40]1.19-1.18 [-2.37, -0.02]1.91 [1.23, 2.62]33% [9%, 55%]62%92%-0.67-0.222026-08-07 → 2026-08-30
tmin12Zbilinear365d29 1 QC1.54 [1.00, 2.12]1.38-0.77 [-2.05, 0.50]1.84 [1.21, 2.37]34% [14%, 56%]66%93%+0.31+0.472026-01-07 → 2026-08-30
tmin12Zbilinearall29 1 QC1.54 [1.00, 2.14]1.38-0.77 [-1.91, 0.57]1.84 [1.19, 2.38]34% [14%, 54%]66%93%+0.31+0.472026-01-07 → 2026-08-30
tmin12Znearest30d24 1 QC1.75 [1.55, 2.04]1.06-1.33 [-1.87, -0.66]1.96 [1.76, 2.24]25% [11%, 36%]58%92%-0.83-0.092026-08-07 → 2026-08-30
tmin12Znearest90d24 1 QC1.75 [1.56, 2.15]1.06-1.33 [-2.00, -0.58]1.96 [1.76, 2.33]25% [4%, 36%]58%92%-0.83-0.092026-08-07 → 2026-08-30
tmin12Znearest365d29 1 QC1.85 [1.62, 2.27]1.03-1.50 [-2.21, -0.86]2.05 [1.82, 2.40]24% [10%, 35%]52%90%+0.17+0.612026-01-07 → 2026-08-30
tmin12Znearestall29 1 QC1.85 [1.63, 2.31]1.03-1.50 [-2.23, -0.90]2.05 [1.82, 2.44]24% [10%, 35%]52%90%+0.17+0.612026-01-07 → 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 6, 00Z bilinear tmax+20+100-10-202026-08-07: -3.87 °F2026-08-08: -1.17 °F2026-08-09: -6.55 °F2026-08-10: -8.88 °F2026-08-11: -9.46 °F2026-08-12: -10.17 °F2026-08-13: -3.46 °F2026-08-14: -6.42 °F2026-08-15: -4.83 °F2026-08-16: -6.78 °F2026-08-17: -8.78 °F2026-08-18: -6.22 °F2026-08-19: -0.76 °F2026-08-20: -5.69 °F2026-08-21: -4.61 °F2026-08-22: -3.62 °F2026-08-23: -9.48 °F2026-08-24: -6.23 °F2026-08-25: -2.73 °F2026-08-26: -8.7 °F2026-08-27: -3.14 °F2026-08-28: +3.02 °F2026-08-29: -4.81 °F2026-08-30: -4.93 °F2026-08-072026-08-30°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 (24 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-4.93
2026-08-29-4.81
2026-08-28+3.02
2026-08-27-3.14
2026-08-26-8.70
2026-08-25-2.73
2026-08-24-6.23
2026-08-23-9.48
2026-08-22-3.62
2026-08-21-4.61
2026-08-20-5.69
2026-08-19-0.76
2026-08-18-6.22
2026-08-17-8.78
2026-08-16-6.78
2026-08-15-4.83
2026-08-14-6.42
2026-08-13-3.46
2026-08-12-10.17
2026-08-11-9.46
2026-08-10-8.88
2026-08-09-6.55
2026-08-08-1.17
2026-08-07-3.87

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KSFO, lead day 6-20 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: 5 days-8 to -6 °F: 5 days-6 to -4 °F: 5 days-4 to -2 °F: 9 days-2 to 0 °F: 3 days0 to 2 °F: 1 day2 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| 4.71 · P90 |e| 8.94 °F-200+20 °Fn = 30 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
-10.00-8.005
-8.00-6.005
-6.00-4.005
-4.00-2.009
-2.00+0.003
+0.00+2.001
+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
2026-08245.60 -5.34
2026-0162.37 -2.21

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-07 → 2026-08-30 30
GraphCast (IFS init)2026-01-07 → 2026-08-30 30
GraphCast (IFS init)2026-01-07 → 2026-08-30 29
GraphCast (IFS init)2026-01-07 → 2026-08-30 29

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 24 +0.74 [0.14, 1.10] yes
NCEP GFS 24 +0.95 [-0.56, 2.09] no
ECMWF IFS HRES 19 -0.82 [-2.31, 1.39] no
Persistence (baseline) 24 +1.85 [-1.02, 3.40] no

Cite this page

ShortCastCheck, KSFO graphcast_ifs lead 6, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KSFO/model/graphcast_ifs/lead/6/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KSFO (San Francisco Intl), model graphcast_ifs, lead day 6. 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/6/

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