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

Chicago O'Hare · GraphCast (IFS init) · lead day 0

KORD · graphcast_ifs · lead 0

Target day = initialization date + 0 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
tmax00Zbilinear30d28 1 QC3.54 [2.81, 4.25]1.49-3.54 [-4.25, -2.81]4.02 [3.21, 4.69]11% [0%, 21%]21%36%——2026-08-01 → 2026-08-30
tmax00Zbilinear90d28 1 QC3.54 [2.85, 4.36]1.49-3.54 [-4.36, -2.85]4.02 [3.26, 4.74]11% [0%, 24%]21%36%——2026-08-01 → 2026-08-30
tmax00Zbilinear365d34 1 QC3.49 [2.87, 4.13]1.43-3.49 [-4.13, -2.87]3.93 [3.24, 4.53]9% [0%, 20%]24%35%——2026-01-01 → 2026-08-30
tmax00Zbilinearall34 1 QC3.49 [2.87, 4.17]1.43-3.49 [-4.17, -2.87]3.93 [3.26, 4.55]9% [0%, 20%]24%35%——2026-01-01 → 2026-08-30
tmax00Znearest30d28 1 QC3.54 [2.79, 4.26]1.47-3.54 [-4.26, -2.79]4.00 [3.17, 4.68]11% [0%, 21%]21%39%——2026-08-01 → 2026-08-30
tmax00Znearest90d28 1 QC3.54 [2.83, 4.39]1.47-3.54 [-4.39, -2.83]4.00 [3.24, 4.73]11% [0%, 24%]21%39%——2026-08-01 → 2026-08-30
tmax00Znearest365d34 1 QC3.51 [2.89, 4.18]1.42-3.51 [-4.18, -2.89]3.94 [3.25, 4.56]9% [0%, 20%]24%38%——2026-01-01 → 2026-08-30
tmax00Znearestall34 1 QC3.51 [2.89, 4.23]1.42-3.51 [-4.23, -2.89]3.94 [3.26, 4.55]9% [0%, 20%]24%38%——2026-01-01 → 2026-08-30
tmin00Zbilinear30d28 1 QC1.28 [0.84, 1.60]1.21+0.68 [0.18, 1.18]1.75 [1.31, 2.08]54% [37%, 77%]75%86%——2026-08-01 → 2026-08-30
tmin00Zbilinear90d28 1 QC1.28 [0.68, 1.77]1.21+0.68 [0.06, 1.41]1.75 [0.93, 2.26]54% [31%, 86%]75%86%——2026-08-01 → 2026-08-30
tmin00Zbilinear365d34 1 QC1.51 [0.97, 2.11]1.50+0.19 [-1.09, 1.04]1.98 [1.43, 2.49]44% [19%, 71%]71%79%——2026-01-01 → 2026-08-30
tmin00Zbilinearall34 1 QC1.51 [0.93, 2.10]1.50+0.19 [-0.76, 1.14]1.98 [1.39, 2.51]44% [20%, 73%]71%79%——2026-01-01 → 2026-08-30
tmin00Znearest30d28 1 QC1.28 [0.94, 1.58]1.28+0.30 [-0.21, 0.83]1.66 [1.26, 1.98]50% [34%, 65%]86%89%——2026-08-01 → 2026-08-30
tmin00Znearest90d28 1 QC1.28 [0.80, 1.71]1.28+0.30 [-0.32, 1.08]1.66 [0.97, 2.13]50% [30%, 69%]86%89%——2026-08-01 → 2026-08-30
tmin00Znearest365d34 1 QC1.58 [1.04, 2.26]1.56-0.20 [-1.47, 0.66]2.03 [1.40, 2.69]41% [20%, 61%]79%82%——2026-01-01 → 2026-08-30
tmin00Znearestall34 1 QC1.58 [1.03, 2.24]1.56-0.20 [-1.20, 0.78]2.03 [1.36, 2.67]41% [19%, 62%]79%82%——2026-01-01 → 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 KORD, lead day 0, 00Z bilinear tmax+10+50-5-102026-08-01: +0 °F2026-08-02: -4.14 °F2026-08-03: -6.48 °F2026-08-04: -4.38 °F2026-08-05: -3.62 °F2026-08-06: -5.86 °F2026-08-07: -5.67 °F2026-08-08: -3.57 °F2026-08-09: -1.51 °F2026-08-10: -8.65 °F2026-08-11: -0.95 °F2026-08-12: -3.02 °F2026-08-13: -0.11 °F2026-08-14: -3.94 °F2026-08-15: -2.3 °F2026-08-16: -2.4 °F2026-08-17: -4.2 °F2026-08-18: -1.39 °F2026-08-19: -3.04 °F2026-08-20: -1.57 °F2026-08-21: -2.54 °F2026-08-22: -4.56 °F2026-08-23: -3.75 °F2026-08-24: -4.53 °F2026-08-25: -5.12 °F2026-08-28: -4.44 °F2026-08-29: -2.86 °F2026-08-30: -4.56 °F2026-08-012026-08-30°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 (30 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-4.56
2026-08-29-2.86
2026-08-28-4.44
2026-08-27—
2026-08-26—
2026-08-25-5.12
2026-08-24-4.53
2026-08-23-3.75
2026-08-22-4.56
2026-08-21-2.54
2026-08-20-1.57
2026-08-19-3.04
2026-08-18-1.39
2026-08-17-4.20
2026-08-16-2.40
2026-08-15-2.30
2026-08-14-3.94
2026-08-13-0.11
2026-08-12-3.02
2026-08-11-0.95
2026-08-10-8.65
2026-08-09-1.51
2026-08-08-3.57
2026-08-07-5.67
2026-08-06-5.86
2026-08-05-3.62
2026-08-04-4.38
2026-08-03-6.48
2026-08-02-4.14
2026-08-01+0.00

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KORD, lead day 0-10 to -9 °F: 0 days-9 to -8 °F: 1 day-8 to -7 °F: 0 days-7 to -6 °F: 1 day-6 to -5 °F: 3 days-5 to -4 °F: 9 days-4 to -3 °F: 8 days-3 to -2 °F: 4 days-2 to -1 °F: 5 days-1 to 0 °F: 2 days0 to 1 °F: 1 day1 to 2 °F: 0 days2 to 3 °F: 0 days3 to 4 °F: 0 days4 to 5 °F: 0 days5 to 6 °F: 0 days6 to 7 °F: 0 days7 to 8 °F: 0 days8 to 9 °F: 0 days9 to 10 °F: 0 daysdashed: P50 |e| 3.69 · P90 |e| 5.51 °F-100+10 °Fn = 34 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
-10.00-9.000
-9.00-8.001
-8.00-7.000
-7.00-6.001
-6.00-5.003
-5.00-4.009
-4.00-3.008
-3.00-2.004
-2.00-1.005
-1.00+0.002
+0.00+1.001
+1.00+2.000
+2.00+3.000
+3.00+4.000
+4.00+5.000
+5.00+6.000
+6.00+7.000
+7.00+8.000
+8.00+9.000
+9.00+10.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08283.54 -3.54
2026-0163.23 -3.23

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-01 → 2026-08-30 34
GraphCast (IFS init)2026-01-01 → 2026-08-30 34

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 28 +0.27 [-0.17, 0.82] no
Aurora (GFS init) 2 -1.39 [-2.65, -0.13] yes
Aurora (IFS init) 3 +1.00 [0.60, 1.79] yes
FourCastNet v2 (GFS init) 2 +2.19 [1.56, 2.82] yes
FourCastNet v2 (IFS init) 3 +0.59 [-0.55, 2.21] no
NCEP GFS 28 +1.24 [0.60, 1.95] yes
GraphCast (GFS init) 2 +0.20 [0.07, 0.32] yes
ECMWF IFS HRES 22 +0.82 [-0.06, 1.73] no
Pangu-Weather (GFS init) 2 +1.35 [0.30, 2.40] yes
Pangu-Weather (IFS init) 3 +1.24 [1.14, 1.45] yes

Cite this page

ShortCastCheck, KORD graphcast_ifs lead 0, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KORD/model/graphcast_ifs/lead/0/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KORD (Chicago O'Hare), model graphcast_ifs, lead day 0. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KORD/model/graphcast_ifs/lead/0/

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